Yes, it is possible to perform correlation or regression analysis between two variables measured on different scales, such as a 5-point Likert scale and a 7-point Likert scale. However, you need to be aware of some potential issues and considerations when working with variables on different scales.

Regarding publications on this subject, there may not be a single publication that directly addresses the combination of different Likert scales in correlation and regression analysis. However, you can find resources on dealing with Likert scale data in general, as well as how to handle variables on different scales in statistical analysis. You may want to explore textbooks and research articles on applied statistics and survey research methods for comprehensive guidance. A few key concepts to look for in these resources are "ordinal data analysis," "scale transformation," and "comparative analysis of Likert scales."

Before doing any numerical calculations here, it would be sensible to just do a visual assessment using cross-plots. These would give an immediate indication of any concerns.

Going beyond the cut and paste answer from Chat-GPT, it is important to know whether you just have a pair of items, for whether you have multiple items with either 5 or 7 response points. If you have multiple items, then you can use coefficient alpha to determine whether it is possible to form interval-level scales by summing the items in each set.

If you have just two items, then you should investigate the various coefficients for calculating a correlation between ordinal variables.

yes، it is possible because you have to measured the score of both construct and then you can use it for further statistical measurements such as correlation or regression۔

Two different scales can often be better compared with normalization or standardization. That is, different scales can make one effect look larger. For example, comparing a 1-7 scale with a 38-100 scale can be very misleading without considering min/max and distribution and degrees of freedom. Outliers can be problematic.

https://inomics.com/blog/standardizing-the-data-of-different-scales-which-method-to-use-1036202

David Coker It seems like your answer its about regression analysis ("effect size"), but every regression program I have ever seen gives standardized as well as unstandardized coefficients, so there is no need to do a separate standardization prior to the analysis.

I agree. Except when cleaning data and examining prior to running analysis, data on different scales are difficult to compare. My example is a real-world problem of two very different scales. For example, straight percentages and graphs that are not standardized/normalized can be tough to eye. So much so, I've seen many people get it wrong (including in my example).