Assessment

Spatial Data Analysis with R (01:450:320)

Overall Grades

Final letter grades will be assigned as follows

lower 90 85 80 75 70 60 0
upper 100 90 85 80 75 70 60
grade A B+ B C+ C D F

Participation

Participation is worth 10% of your final grade and has two components.

  • Attendance. You are expected to attend each class. If you cannot make it for some reason, please arrange ahead of time.

  • Engagement with the course material, including any coding work assigned for homework. It is essential to read and work through the class material between classes. You will not be able to learn everything just from class. Additionally, you will at times be asked to work on practical problems between classes, for which you should be prepared to share your code and provide examples of errors/obstacles.

  • Reading quizzes. Periodically, short quizzes may be given on core concepts from readings. These quizzes will be closed-book and not exceed 10 minutes. (e.g. explain in general terms how an apply function works). Students receiving 60% or higher on a quiz receive a “Complete” grade.

Assignments

Assessment: Can we load the data you used, run your code, and arrive at the same result you reported? Does your code conform to the style guide (note: we are a little obsessive about this)

We will grade using this rubric (or one close to it that captures the results/style/reproducibility dimensions).

Accuracy APts Reproducibility RPts Elegance EPts
Results all wrong (0-5%) 0 Can’t be made to run 0 Indecipherable code, no comments/documentation 0
Results mostly wrong (5-25%) 4 Doesn’t work 4 Indecipherable code, no comments/documentation 2
Results more wrong than right (25-50%) 8 Doesn’t run without numerous fixes 8 Code indecipherable/no comments/documentation 4
Results more right than wrong (50-75%) 12 Runs after a handful of fixes 12 Code readable but wrong style/some commenting/doc’tation 6
Results almost entirely correct (75-95%) 16 Runs after some minor tweaks 16 Code mostly conformant/adequate commenting/documentation 8
Results all correct (95-100%) 20 Runs out of the box 20 Code conformant and stylish/well commented/documented 10

⚠️ Important: Assignment grades are not final until the corresponding quiz is completed. The quiz score defines the maximum possible points you can receive for that assignment.

Quiz-Based Grade Refinement

Each assignment is followed by an in-class, closed-book quiz consisting of 10 questions directly related to the programming tasks you just completed. Your final score for the assignment will be determined by your quiz performance according to the following tiers:

Quiz Score Assignment credit Description
8-10 100% Mastery: Demonstrates clear conceptual understanding.
6-7 85% Proficient: Demonstrates general understanding, with minor gaps in specific technical details.
4-5 70% Developing: Significant gaps in understanding.
0-3 60% Deficient: Inability to explain core components of the submitted work.
No Quiz 0% Unverified: No credit is awarded if the validation quiz is missed.

Final project overview

For the project overview assignment, assessment will be out of 50 points using simplified and somewhat different categories than the unit assignments, as shown below. The required content for this particular assignment in the Final Projects section.

Note that the minimum number of points assigned per category assume that the final project overview was completed and handed in.

Quality of Plan QPts Structure/completeness CPts Reproducibility RPts
No plan (analysis/coding approach and timelines) provided 5 Doesn’t follow required structure or provide requested information 2 Can’t be built/no R package structure 2
Plan (analysis/coding approach and timelines) is provided but mostly unworkable, many holes or inconsistencies in logic or approach 15 Barely follows required structure or provides requested information 4 Runs after many fixes 4
Probably workable plan, but with a fair number of holes or inconsistencies that need to be addressed 20 Partially follows structure and provides part of the information 6 Runs after 4-6 fixes 6
A pretty well thought-out and workable plan, with just a handful of holes or inconsistencies 25 Mostly follows structure and provides most of the required information 8 Runs after 1-3 fixes 8
Excellent, well thought plan, no obvious flaws in approach 30 Perfectly follows structure and provides all of the required information 10 Runs out of the box/with no fixes needed 10

Final project oral defense

The Oral Defense is a 10–15 minute technical briefing designed to validate the project plan submitted in your written report. This session serves as a “sanity check” to ensure your proposed workflow is technically sound and that you/your group have a complete command of the logic and tools required to execute the analysis.

Assessment is out of 50 pts, and will be based on three categories:

  • Verification: To confirm that the project architecture and code selection were personally developed by the student/group.

  • Feasibility: To ensure the proposed packages and functions are appropriate for the research questions involved.

  • Clarity: To provide an opportunity for the student to articulate the “analytical bridge” between raw data and expected research outcomes.

Note that the minimum number of points assigned per category assume that the student or group is present for the defense.

Question handle QPts Technical literacy LPts Logic & Outcome OPts
Unable to answer any questions; suggests a significant lack of preparation 4 Namedrops packages but unable to explain what they do or why they are needed 4 Unable to explain how the proposed analysis will lead to the intended result 2
Responses are vague, inconsistent with the report 8 Understands functions and packages but struggles to explain some of them 8 Describes outcomes that will not logically follow from the proposed analysis 4
Answers basic questions but struggles with technical “why” or “how” details 12 Understands general package use but cannot describe the specific necessity of functions 12 Describes expected maps/stats but struggles to connect how they will answer the research question 6
Answers most questions well; shows clear familiarity with only minor hesitations 16 Accurately describes the role of specific functions and the packages they intend to use 16 Explains the step-by-step logic of how processing will lead to a research result 8
Answers all questions thoroughly and fluently 20 Explains the underlying logic of chosen functions and justifies why alternatives were rejected 20 Articulates exactly what results will look like and how they will validate the thesis 10

Final project

Assessment is out of 70 pts (note that there a minimum points floor, which assumes that a final project has been submitted), and will be based on four categories:

  • Quality evaluates the content of the material in your package, focusing on the information and ideas conveyed in your descriptions, figures, and tables. How well did you understand and execute the project, and how well did you convey what you did?
  • Clarity examines the clarity of the writing (is it easy to understand, no typos or missed words), the visual aspects of figures (does the color choice make sense, legend sensible, size appropriate, text readable?), formatting (did html lists end up as lists? do headings make sense?), and code syntax. This basically covers the stylistic components of your vignettes and code.
  • Reproducibility relates to the now familiar criteria about whether your package builds for another user without error, and with browsable vignettes.
  • Progress evaluates whether you achieved the objectives put forward in your Assignment 6 and final presentation, focusing more on the latter due to course corrections (and incorporating any necessary changes we discussed)

Note that the minimum number of points assigned per category assume that the final project was completed and handed in.

Quality QPts Progress PrPts Clarity CPts Reproducbility RPts
Objectives/methods/results/interpretations unclear, poor understanding conveyed replete with analytical/methodological errors Findings/interpretations/examples absent 10 None since presentation 5.0 Vignette(s) a complete mess, graphics absent/totally confusing, text a word salad, narrative impossible to follow code a mess/doesn’t follow style 5.0 Can’t be built/no R package structure 3.0
Objectives/methods/results/interpretations cloudy, understanding of project less than half many analytical/methodological errors Findings/interpretations/examples present but weak/wrong 15 Only a little 7.5 Vignette(s) more unclear than clear; graphics present but hard to decipher; text somewhat confusing; narrative hard to follow code messy/mostly doesn’t follow style 7.5 Runs after many fixes 5.0
Objectives/methods/results/interpretations 50% clear, project about 50% understood some analytical/methodological errors Findings/interpretations/examples about half wrong 20 Modest progress 10.0 Vignette(s) not unreasonable; graphics so-so, not spectacular, maybe hard to decipher; text still a bit distracting, causes some confusion; narrative mostly clear code sloppy/half follows style 10.0 Runs after 4-6 fixes 7.0
Objectives/methods/results/interpretations 75% clear, project about 75% understood few analytical/methodological errors Findings/interpretations/examples interesting/correct 25 Project mostly complete 12.5 Vignette(s) good; graphics mostly clear, no major questions; text understood and gets the job done; narrative clear code clear/mostly follows style 12.5 Runs after 1-3 fixes 8.5
Objectives/methods/results/interpretations crystal clear, project completely understood no obvious analytical/methodological errors Findings/interpretations/examples novel/really well done 30 Much greater than expected 15.0 Vignette(s) outstanding; graphics spectacular, tell a clear and easily understood story supporting text; text crystal clear and right amount; gripping narrative code tight, perfect style 15.0 Runs out of the box/with no fixes needed 10.0