Hello and welcome to the seventh edition of empirical research methods and medical

engineering.

Today we'll talk about data visualization foundations of statistics and explorative

data analysis.

Could someone just give me a heads up whether the recording started and whether you can

see the slides, you can write in the chat.

Yes, so we did.

Great.

Thanks Frank.

But yeah.

Okay.

Yeah, I'm sorry.

So first we'll talk about data visualization today.

I'll show you some different kinds of visualizations that can be done on one example data sets.

Then we'll talk about foundations of statistics where I have to tell you right away that unfortunately

I could not prepare as well as I would want to do so I cut a little part out that we

will then do afterwards, especially on correlation covariance.

So I just have a little introduction so to say that will help you through the practical

lesson that will be there later on.

But I'd like to tell you some more and I will do that next time.

And then we'll talk some more about explorative data analysis.

Okay, but first to data visualization.

Sorry.

So we have this graph here.

And I mean, all of you have seen a graph and a question to you.

What does it tell you?

What can you read from that and that might be interesting or of an issue with the visualization?

Do you have any ideas?

You can also write in the chat if you feel more comfortable then because then you won't

be recorded and I will read it out loud.

Or I have an answer prepared of course, but I just thought maybe you will have this little

interactive part in the beginning.

Okay, it's fine.

Then I'll speak out.

It's okay.

So what we see here is so we have here, especially in this graph showing male and female students

in thousands over the years and how they signed up for studies.

So we see here in light gray, we see the male students in dark gray, we see the female

students and this line shows us the line of how many were female students of 100 students.

And what you see is that although this line, some does follow maybe some kind of trend

that was also shown in the female students, it's definitely not the same and it actually

in some parts.

So like for example, this bump here that also goes for the male students is not reflected

there.

So it kind of changes the perception does not answer the question of how many female

students were there.

This one is strongly dependent.

The line is strongly dependent on the male students as well and not just purely on the

female students.

So what it shows basically is when you visualize, you should think about what you want to visualize.