search

Dr-Eberle-Zentrum / Data-projects-with-R-and-GitHub

6 stars 3 forks source link

Comments for Arlene by Sarah #248

Closed sarahloeber closed 6 days ago

sarahloeber commented 3 months ago

Hi @ArleneHohl,

very cool project idea! You’ve made it clear which question you want to answer and it’s very handy you’ve included the codebook.

As for the dataset and project description, it would probably help to define the countries you want to look at beforehand. This would not only make the size of the dataset more manageable immediately as you can exclude other countries, but also refine the topic a bit more.

Maybe you could give a little more detail about the score of the education qualifications the students expect to achieve. You said you need to build a score- so it would be the same procedure as for an aggregated score of the socioeconomic status? I was looking through the codebook but couldn’t quite make sense of that.

I also really like your idea about plotting the OECD average for every country in the background. In the plot you added, this would have indeed made the graphic more comparable. Perhaps it is an idea to use different labels on the x-axis and not quintiles. I personally find that a bit confusing, but maybe that’s just me.

Is it possible to add this plot as an image in your RMarkdown file? That would also help making the project more “polished”.

All in all, this is a nice achievable project 🙂

Best, Sarah

ArleneHohl commented 3 months ago

Hi Sarah,

thanks a lot for your feedback!

Yes you are right, I should definetely choose the countries to make the dataset more manageable. At first my plan was to identify the countries with the highest, lowest and the one with mean socioeconomic status. The problem is just that you then have to work with the big dataset first. @martin-raden had the idea we could just use the three countries named in my exemplatory graphics to make things easier. What do you think about that?

About adding the OECD mean: I think I might add this as an optional task, as this would require working with the big dataset as well...

Yes I feel so as well, the quintiles are quite confusing. Would you rather make a scatterplot instead? Sure, I will add the plot into the Markdown, thanks for the comment.

Best, Arlene