More Bananas = More Friends – Causation VS Correlation

Correlation and Causation. Two big words that both start with the letter ‘C’. But what do they really mean? In the second Scimatics project of grade 9, we’ll be conducting a survey, to see how correlation and causation are different. 

At the start of the project, we did a project mind map to see what knowledge I already knew, and to see what questions I have about the up coming project. This was good because I knew virtually nothing about these two concepts.

  • Knowledge Before the Project

Planning and Conducting:

For this project, we had to do a presentation using survey results that we gathered, to show two correlations, and one causation. To achieve this we started working on the questions that we were going to ask in our survey. The hard part about making the survey was asking the right questions. We all had predictions on what the answers were going to be, but it wasn’t until after that we would see if we would have any correlations. Because of this, asking the right questions before hand to maximize the chance of finding correlations, while still asking interesting questions was very important. 

Organizing our Questions

Applying and Innovating:

While writing the survey, it was also a challenge to follow good survey ethics while writing the survey. A lot of the most interesting statistics would also not be appropriate to ask people in a survey, and could cause them to lie about the answers. Trying to make sure that all the questions were appropriate to ask and that people wouldn’t lie about them was a challenge. It was also hard to reduce the about of biases in the survey. Things like how the question is worded, what time people take the survey, and many other variables can all influence and trick the audience in answering a certain way, however I think overall we did a good job on this. My partner and I spent a lot of class time trying to make sure the questions were as clear as possible. There were some questions that could have needed a bit more clarification, but over all I don’t think there were many biases in the questions themselves.

Speaking of partners, this was a partner project. My partner was Alicia, if you want to see her perspective on the project you can check out her blog here.

Monkey’s Guide to Alicia 

After that we had to format all of the data into one giant table to make it as easy as possible to find any correlations. Luckily, we managed to find quite a lot of correlations, some with causation and some without. With this we planned a presentation to explain our findings. 

All of Our Data

Communicating and Representing:

I think what I most liked about our presentation was the role that the keynote played. I’ve often struggled with balancing the information that goes into the keynote, and the information that we will talk about. I think this presentation had really pretty good balance between the two. The keynote amplifies what we are saying. It can’t be used on it’s own, the two mediums need to work together.  You would probably notice that if you just watch the video. There is a lot of information left out, but I think that’s a good thing because that makes the information that is on the presentation more readable and easier to understand. That being said, making the graphs also not misleading to the audience also took some thinking to get right. Overall I think that we showed off and explained our correlations and causations in a very clear and interesting way in the the presentation.

Finally for the driving question “How are Correlation and Causation different”. Well, correlation is the connection or relation between two different data points. A common example of this is the correlation between the amount of people getting sunburn and the amount of ice cream people consume. In this case, both factor A (People with sunburn), and factor B (Ice-cream consumed) are related, but one does not have to directly affect the other. The difference with causation is that one factor, has to directly affect the other. An example of this could be that the taller you are, directly affects how long your wingspan is. This means that all causations also have a correlation, but not all correlations have causation, which can be confusing and exploited in ways that make it seem like there is causation where there really isn’t any. 

Anyways thanks for reading. I enjoyed this project and trying to find as many wacky correlations that we could.

See you next time,

📈Nolan📉

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>