Maybe you also want to look at construct and convergent validity as well.

Assuming that the confidence intervals of HTMT estimated by bootstrapping reported that the confidence intervals did not include 1 in any of the cases you are testing, then the discriminant validity holds for the measurement model.

Dear Sara Hoseingholizade

In my opinion, the three measures for the discriminative validity should hold (cross-loading, Fornell-Larcker and HTMT) to ensure that the measure validity is out of the question. Therefore, If I were you, I would eliminate them or if possible, I would assign them to another construct.

Best Regards,

Belal Edries

Thanks, Pradeep Paraman. Since Hair (2017), has mentioned, HTMT is an alternative method for discriminant validity. Still, do you know any other specific research or source supporting it?

Thanks, Belal Edries ,

Ideally, I myself would prefer all three criteria to meet standards, but they don't!

Eliminating or merging constructs are also impossible in this model.

Dear Sara Hoseingholizade

Regarding your items that have cross-loading issues, I would like to draw your attention to a set of actions that may apply and get your measurement model through the validation process.

• You probably have applied this before but I would suggest rechecking the data, for any outliers that might be still there.
• Check your sample size with relative to the number of items you have in your study. Roughly speaking, if you have between 10 to 20 respondents per item, that should be okay. However, if possible aim to collect more response.
• Drawing the factor correlation circle would be somewhat useful to present you a graphical image of the current situation, however, unfortunately, what I had is a code for drawing the correlation circle of the items using PCA using R software but I have included it anyways with a brief illustrating of each line purpose.
• In some cases, the miswording of the questions may lead to some degree of pushing one of the factors in another direction thus inverting the scale is usually applied to overcome this issue. Therefore, I would suggest rechecking the question and ensure that if the respondents read the questions carefully and they were designed to measure the constructs reflectivity, then the responses should be in one direction. If you notice that many respondents have not met this criterion then probably should consider to take an action.
• This might the least favourable action, however, if applicable to put these two items together with the construct that have cross-loaded. Here is an example, assuming that construct A(3 items) and construct B(3 items). In your situation, you had 2 items of B showing more load on A, thus you may apply that A(5 items) and B(1 item). This action is likely to reduce the reliability of the study, however, it is still a possibility.

Throughout my research, I have noticed that only a few studies have applied HTMT. Here is one of them:

• Benleulmi, A. Z., & Blecker, T. (2017). Investigating the factors influencing the acceptance of fully autonomous cars. In C. M. Kersten Wolfgang Blecker, Thorsten Ringle (Ed.), Digitalization in Supply Chain Management and Logistics: Smart and Digital Solutions for an Industry 4.0 Environment. (Vol. 23). epubli GmbH. https://doi.org/10.15480/882.1449

Best Regards,

Belal Edries

Values below .9 indicate evidence of discrimination validity.

See example: Article A Generational Study of Employees' Customer Orientation: A M...

Dear Belal Edries ,

Thanks a lot for the detailed answer. I guess I have to read the questionnaire all over again and see if there can be misinterpretations for questions.

Thanks for the R code, I will check the factor correlation as well.

Bests,

Sara

Thanks Ali B. Mahmoud . I will check it.

It is really okay. No harm to do what you have done.

Fornell & Lacker is backed (Henseler et al, 2015). rather avoid.

Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115-135.

for discriminant validity, HTMT

Thanks Md Karim Rabiul & Mahadi Hasan Miraz

Sara Hoseingholizade You are wellcome. Keep it up.

Hi all,

What I have found so far are:

In assessing the discriminant validity, the HTMT criterion is superior to both the Fornell and Larcker criterion and the assessment of (partial) cross-loadings (Henseler et al., 2015).

Although the Fornell–Larcker criterion [96] had been long recommended to assess discriminant validity of latent variables [12], it is ineffective in combination with traditional PLS-PM because it relies on consistent factor loading estimates. HTMT is a reliable tool for assessing discriminant validity, whereas the Fornell–Larcker criterion has limitations that do not justify its reputation for rigor and its widespread use in empirical research (Henseler et al., 2020).

Ringle himself in his paper "Elderly customers’ reactions to service failures: the role of future time perspective, wisdom and emotional intelligence" has only used HTMT for discriminant validity.

According to these papers, I believe HTMT suffice for establishing discriminant validity.