ACT B - Evaluating & Comparing Models: MAT135 - Calculus 1 - Fall 2024
ACT B - Evaluating & Comparing Models
You are a Research Assistant with the UN and you’ve just submitted your first modeling assignment. Now that you’ve done that, your supervisor wants you to analyze your work for your progress report. They also want you to work on your speaking and your verbal communication skills. There is another new RA who has just done the same thing, so you are teaming up to complete the analysis together. Since a lot of reading is already required in their job, your supervisor has requested that the reflection come in the form. of a short video. Your goal is to put together a video presentation analyzing the results of both modeling projects.
Audience.
Your audience is your supervisor at the UN. You can assume that they have the same skills and knowledge as you do.
Steps to complete project
Follow these steps to complete the project
Step 0.
Find a partner: you will meet your partner in your tutorial during Week 7. We recommend that you find someone who has completed a report with the same indicator as you.
(Technical Details: You can be researching the same indicator OR the same country, but cannot be researching the same indicator AND the same country.)
Step 1.
Gather your team's two reports from ACT A. Read each other's reports.
Step 2.
Meet with your partner (a virtual meeting is okay!) and discuss the following questions.
Discussion questions:
1. What real-world features of the indicator or country (like war or labour strike) influence how the data changes, among the data points that you collected?
2. How did the different factors and issues that you discussed in (1) impact the your model from ACT A?
3. What additional information and/or data would you need to be able to construct a good model (i.e. better than what you did in ACT A, like what a professional instead of AI might produce)? How would you find that information?
4. When you graph your data, does it make a similar shape as your partner’s data? How does this influence the type of model (i.e. exponential, polynomial, etc) that you might decide to use?
5. In summary: What features would a “good” model have if you were going to make one for your indicators, and is it the same for both indicators/countries? Reflect on what you’ve learned about the modeling process, and what you would hypothetically need to do to perfect your ACT A.
Here are the types of things that you may include with an accompanying explanation for Question 5 (you won’t have time for them all, and you might include other questions too).
How we define the problem (what exactly are we modeling?)
Reasonable & useful assumptions
Mathematical complexity/simplicity of the model
Reliability of data used to create the model
Considering the limitations of a model, e.g. a percentage can’t be over 100%, so a linear model (that isn’t ‘capped’ somehow) isn’t good for modeling percentages
Step 3.
Write a script. for the video based on your discussion. The script. is the draft that you will submit for peer review during your Week 8 tutorials (due Nov 7, 8:59 am). Only one group member submits the assignment to gradescope.ca. After submitting, make sure to add the other group member on gradescope (it will be a “group assignment”).
The Script. should satisfy the following requirements:
Be about 250-500 words long (for 2-4 minute video)
Refer to graphs of your data
Includes a very brief introduction in the video, with your indicators and countries mentioned
Includes the following information in the discussion:
Information and/or data you would need to be able to construct a good model & how you might find that information
Some real-world features or issues in the country(ies) or with the indicator(s) that influence how the data is changing (discuss for both models, as separate data sets)
How the data changes in the two countries (use words like increasing, decreasing, slower, faster, speeds up, slows down, etc.)
Includes a conclusion that reflects on how the differences in real-world factors change your indicators in different ways, and, in turn, how these differences would result in differences between the mathematical models that you would choose for your data, assuming you did not have to use AI (you only need to discuss the type of model, do not go further in the modelling process).
Tells each speaker when they will talk, and provides them with approximately equal speaking time
Your audience is now different, so you should speak in a different way to them than you did in your general audience piece before
Step 4.
Record video presentation (click here for instructions on how to record it (https://q.utoronto.ca/courses/353446/pages/zoom-recording-instructions) ).
The video should satisfy the following requirements (in addition to following the script. above):
Be 2–4 minutes in length
(we recognize that English Language Learners might speak more slowly – that's okay! If this applies to you, look at the script. word guidelines of 250-500 words. However, this time limit should be considered “strict” by those who are not English Language Learners.)
Feature both audio and video of high-enough quality to be viewed easily
Include a presentation with relevant graphs and/or visualizations
Include both speakers' faces
Does not include music
Posting /Submission guidelines will be added
Submitted through gradescope.ca
Suggestions:
The following are suggestions. They are not required, but they will likely improve your ACT.
Be sure to record horizontally so that the picture fills the whole screen.
You will need someone else to hold the camera for you if you are not using a tripod or a laptop camera.
Tip: you can record a Zoom session if you are not able to be in the same physical location, but you will want to practice first!
Make sure that the presentation graphs and text are clearly readable in the video.
When you are done, check the video for clear audio and video quality.