Soumaya Elhassouni. My general sense is that your student should be encouraged to adapt the previous scale because it's best to use scales that measure what researchers want to measure rather than scales that measure something else.

However, there could be some problems. For example, does the original scale have copyright on it (in which case its features might have limited transferability)? And might your student want to compare the new results with the results of other researchers?

Furthermore, creating new scales can involve a lot of work if the process is to be done competently. Perhaps it would be necessary to check with an "expert" group to ensure that the new items are relevant for the target population. Then, once the items have been determined, factor analysis should be used to see what the structure of the items / scale is. For that, at the initial stage, exploratory factor analysis would be most appropriate, but that requires an adequate sample size, which could be difficult to obtain.

Wish your student all the best on my behalf!

To the best of my knowledge,it is possible to add new items to the original scale, but you need to have strong justification on it.

As Dr. Robert Trevethan mentioned can adopt the panel of "expert" for the validity analysis (face validity and content validity) which scrutinize to find out if the item(s) predicted to measure what they should exactly measure. In this step for the "necessity of items " you can use CVR (content validity ratio) proposed by Lawshe (1975)1. Then, Relevancy of items can be measured by content validity index (CVI).

If item (s) passed the validity step and not excluded, to determine the level of reliability, can apply the inter-item correlation and Cronbach’s alpha.

* A correlation value less than 0.2 or 0.3 indicates that the corresponding item does not correlate very well with the scale overall and, thus, it may be dropped (Field, 2009)2.

**According to Sekaran (2006)3 if the value of Cronbach’s alpha reliabilities is less than 0.6, it would be considered as poor, if the value is in 0.7 range, it is acceptable and the reliabilities value above 0.8 are considered good.

Good luck

1)Lawshe, C. H. (1975). A quantitative approach to content validity . Personnel psychology, 28(4), 563-575.

2)Field, A. (2009). Discovering statistics using SPSS. London: Sage publications.

3)Sekaran, U. (2006). Research Methods for Business: A Skill-Building Approach. (4th, Ed.) New Delhi: John Wiley & Sons.

I agree with the above, I must have a theoretical justification that corroborates the sum of the proposed factors of the variable. Performing the AVE with factorial loads greater than .70 for convergent and discriminant validity with other instruments that evaluate the same variable and even the original using the HTMT method can demonstrate that beyond the correlation there is a theoretical and methodological discrimination of the instrument.

Alireza shabani shojaei. Thank you for adding to what I had suggested. Interestingly, I have just submitted an article in which we used the CVR. Here is an article that we used and that I think might be more helpful than is the Lawshe (1975) article:

Wilson, F .R., Pan, W., & Schumsky, D. A. (2012). Recalculation of the critical values for Lawshe’s content validity ratio. Measurement and Evaluation in Counseling and Development, 45, 197–210. https://doi.org/10.1177/0748175612440286

On something else that you referred to, for what it's worth, I think there is a fair amount of evidence that coefficient alpha (often referred to as Cronbach's alpha) is a considerably more complex metric than is often believed. For example, high alpha values can be indicative of undesirable redundancy among a set of items, and they can also be pretty much an inevitably if a large number of items are involved (even if many of those items don't relate to many of the other items).

I hope that's helpful.

Robt.

Robert Trevethan

Thank you for sharing the article, soon I will read it.

Best,

Alireza

Alireza shabani shojaei. You're welcome. If I might be able to help further, pls feel free to ask.

Incidentally, here's a good article about coefficient alpha:

Streiner, D. L. (2003). Starting at the beginning: An introduction to coefficient alpha and internal consistency. Journal of Personality Assessment, 80(1), 99–103. https://doi.org/10.1207/S15327752JPA8001_18

I could give you additional references if you're interested - though soon I hope that my colleagues and I will have a substantial article published about these kinds of things - open access.

Dear Dr. Robert Trevethan

Of course I am interested and I really appreciate your consideration.

I will wait to read more references in this matter.

Thank you so much.

Alireza

Respected Prof. Soumaya Elhassouni I agree with Dr. Robert Trevethan that developing a new scale or adding content (more items) into an existing measurement scale is a tedious job and requires many reliability-validity steps.

Moreover, regardless of the application of statistical procedures (EFA, CFA, CCA, etc.), in general, it is hard for a novice researcher (e.g., a Ph.D. student) to "Redefine" an existing concept and so to add further content dimensions (items) while operationalizing it.

Best regards

Muhammad Zia Aslam. Maybe we are getting away from the original issue, but I agree with you about the amount of effort required to develop a new scale. Actually, I created a new scale as part of my own PhD and it was a mammoth task. I decided to do it, however, rather than use a scale "from the shelf".

My view was that it was better to go to the effort of measuring something that I wanted to measure rather than measuring something else and hoping that it would do the trick despite being slightly off beam.

I am currently doing research that involves burnout and teacher sense of efficacy - and in both of those fields of research I am concerned about the validity of the scales that have been used for at least 20 years. As a result, my colleagues and I are going back to "basics".

Dear all,

Thank you very much for the time and rigor of your answers. Very kind of each of you and very useful. Thanks again

Thankyou everyone here for your very helpful question and responses. I am about to do this in the education field for the third paper of my PhD and your suggestions have been invaluable. I look forward to citing you.