I'm not perfectly clear on your model. Decide if doing each phase one at a time (sequential) or at the same time (called concurrent, simultaneously, or parallel). Will one phase influence the next? If so, in your model, you have exploratory. If both at same time, probably embedded.

See https://www.sagepub.com/sites/default/files/upm-binaries/10982_Chapter_4.pdf&ved=2ahUKEwimwu7-zLv2AhUEAp0JHQx1C8kQFnoECAgQBg&usg=AOvVaw0GFWPm63fPj-hqOIG8gW3A

Also examine https://journals.sagepub.com/doi/pdf/10.1177/1077800410364741&ved=2ahUKEwjAoOSrzbv2AhWBVs0KHT0zBsoQFnoECAcQAQ&usg=AOvVaw1a6lk3NyLoUlyWFP-MGoMx

One more thing: I used John Creswell, but he is not the only framework. There are over 40. You can be very creative as long as the model you adopt answers your research question and provides reliability and validity. Good luck.

First, there is no agreed-upon typology in mixed methods for multi-stage designs. Second, there is no need to follow any particular design typology in mixed methods research, because what matters most is the match between your research goals and your research procedures.

The main concern I would have about what you have outlined is the final stage of integration. If your encoding/decoding approach provides a model for that, then everything is fine. If not, I would urge you design this step before your start, so that you create the other steps in a way that will facilitate integration.

Hi David C. Coker,

Thank you for your answer and links.

The model is sequential, phase one will inform phases two and three in the analyses (merging). It will inform the user's classification. Also, phase two will influence sampling in phase three.

So I am not sure how should I classify the model? and can you suggest a framework that would be more suitable for the process I am following?

Also, I will be grateful if you could suggest some studies that have similar prosses.

Thank you again for your time!

Hi David L Morgan

Thank you for your answer, appreciate it. Your answer made me rethink the variables, and based on your common I re-designed the model with the variables that will be analysed in each stage, and most importantly the integration stage (attached).

Are there any issues with it? and can you kindly suggest any reading that could help me justify the model?

Thank you again for your answer and time!

David L Morgan is right, but it is not either/or. You probably need a justification for your dissertation chair and ultimately committee. While I don't get hung up on names, you are apt to have sources to back up your model. Please see my aforementioned citations, and with the following article, you should be well on your way. You might mine the bibliographies to find specifics.

Article How to Construct a Mixed Methods Research Design

Your revised model seems to say that everything affects everything. I personally prefer your original model. And I think that having a unique model is just fine. There is no need to rely on past precedents.

Clean it up. Clarity is needed. Personally, I think Harvard Catalyst is an easy guide to get you started on your visual. (Which is really based off Creswell.)

See https://catalyst.harvard.edu/community-engagement/mmr/

Agree with answers above

Hi,

Thanks again for your time.

I revised the model and used Harvard Catalyst guide. I removed the variables and will demonstrate them in the chapter. Hope it is clearer now

Check http://komejournal.com/files/KOME_EvanKropp.pdf

https://methods.sagepub.com/reference/the-sage-encyclopedia-of-social-science-research-methods/n279.xml

You can do triangulation by talking to the users, developers and regulators.