You are not going to stop plagiarism by making the writing process transparant. There are other tools for that. It may help to prevent going through many hypotheses until you find one that is "significant". Keeping the write process transparant for this purpose is often discussed under the label pre-registration. Combine that with a clear procedure for making sure that the study can be replicated, and you are well on your way to open science.
There are multiple initiatives in this realm. Often this involves not just writing but also study design, data collection, data preparation and analysis, ... One such initiative is: https://osf.io/ . On that site you can find numerous examples of real research projects.
This may be overkill for bachelors and master projects. For them I would simulate the process by letting them pre-register their project with me. I will be explicit about it, so it is a learning experience. For PhD projects the of site can definitely be useful.