There are 3 graded assignments in the course. These will assess the programming and reasons skills developed in the weeks before the assignment submissions. The contents of the assignments will be released according to the schedule.
- Assignment 01 - Data Collection and Wrangling (15%)
- Assignment 02 - Geographic Visualisation (15%)
- Assignment 03 - Prediction/Inference (15%)
Marking Criteria for assignments
The following rubric will be used for assessments. See the column with points (2) for what is expected of an ideal assignment. Points will be deducted if elements are missing and/or incorrectly done:
|Output||0.5||the code does not produce the expected output||the code partially produces the expected output||the code shows the expected output
- code runs
- code uses reproducible paths
- code has no errors or warnings
- expected output is clearly presented through code/markdown explanations and figures
|Formatting||0.5||code is not formatted at all or messy||code is partially formatted||code is properly formatted
- variables and functions are named meaningfully like “crime_incidence as opposed to c_2513”.
- code is written using functions
- code is written in sections using headers and explanations in markdown
- code is commented
(e.g. Tidy Data, EDA, Graphical Excellence, Spatial Autocorrelation, Network Weights, Regression, etc.)
|2||code/output shows no evidence of understanding of methods presented in class and homework exercises||code/output shows some evidence of understanding of methods presented in class and homework exercises||code/output illustrates clear evidence of understanding of methods presented in class and homework exercises
- models are used correctly
- use of visual inspection to explain model outputs
- axes labels are named meaningfully, legends are present
|Documentation (Markdown/Comments)||2||there is some documentation to follow your code. (e.g. why did you drop a certain variable or why did you choose one over another?)||there is some documentation explaining your code’s logic and your choices in the analysis but these choices or logic is not clear.||there is extensive documentation explaining your code’s logic and your choices in the analysis.
- hypothesis clearly stated
- explanation of model choices clearly presented (ex. variables, number of neighbours or clusters, etc.)
- interpretation of model results, weakness and strengths (ex. errors in regression)
- well substantiated argumentation of results and choices
- plots are clearly described and interpreted
Submit assignments and final project on Brightspace. Formats and naming conventions are given within the respective files.