Section Assessment and Grading
Your project will be assessed based on how well you demonstrate data science skills and concepts across all phases of the work. The grading emphasizes both technical proficiency and communication effectiveness.
Subsection Evaluation Rubrics
Each project component will be evaluated using the following rubrics:
Subsubsection Dataset Selection Report
For an A grade:
- Dataset Suitability: Dataset perfectly meets all selection criteria; rich with potential for analysis
- Documentation: Comprehensive documentation of source, contents, and context
- Initial Exploration: Thorough initial exploration with insightful observations
- Ethical Considerations: Thoughtful discussion of relevant ethical considerations
For a B-C grade:
- Dataset Suitability: Dataset meets most criteria; adequate for project purposes
- Documentation: Adequate documentation but missing some details
- Initial Exploration: Basic exploration with some relevant observations
- Ethical Considerations: Basic mention of ethical considerations
For a D-F grade:
- Dataset Suitability: Dataset fails to meet multiple criteria; limited potential for analysis
- Documentation: Minimal documentation; unclear source or context
- Initial Exploration: Superficial exploration with few or no observations
- Ethical Considerations: No consideration of ethical dimensions
Subsubsection Investigation Plan
For an A grade:
- Statistical Questions: Clear, specific, meaningful questions that anticipate variability and are answerable with data
- Data Assessment: Comprehensive assessment of data quality, completeness, and limitations
- Analysis Approach: Well-reasoned approaches for each question, with clear justification
- Plan Completeness: Thorough plan addressing all aspects of the investigation
For a B-C grade:
- Statistical Questions: Mostly clear questions, but may lack specificity or clear connection to data
- Data Assessment: Basic assessment of data with some gaps
- Analysis Approach: Reasonable approaches but limited justification
- Plan Completeness: Adequate plan but missing some elements
For a D-F grade:
- Statistical Questions: Vague, overly broad, or unanswerable questions
- Data Assessment: Minimal or unrealistic assessment of data
- Analysis Approach: Inappropriate or vague analysis approaches
- Plan Completeness: Incomplete or insufficient planning
Subsubsection Data Processing Documentation
For an A grade:
- Data Cleaning: Appropriate handling of missing values, outliers, and other quality issues with clear rationale
- Data Transformation: Creative and appropriate derived variables, filtering, and other transformations
- Documentation: Clear, comprehensive documentation of all data moves and decisions
- Preliminary Analysis: Insightful summary statistics and group comparisons
For a B-C grade:
- Data Cleaning: Basic cleaning with some rationale, but may miss some issues
- Data Transformation: Basic transformations that support analysis
- Documentation: Adequate documentation but missing some details
- Preliminary Analysis: Basic summary statistics with limited interpretation
For a D-F grade:
- Data Cleaning: Inadequate cleaning or inappropriate handling of data issues
- Data Transformation: Few or inappropriate transformations
- Documentation: Poor or insufficient documentation
- Preliminary Analysis: Minimal or incorrect statistical analysis
Subsubsection Final Presentation
For an A grade:
- Content: Comprehensive coverage of context, methods, findings, and implications
- Data Story: Compelling narrative that effectively communicates insights
- Visual Support: Excellent use of visualizations to support and enhance presentation
- Delivery: Clear, engaging delivery with strong command of material
- Q&A Response: Thoughtful, knowledgeable responses to questions
For a B-C grade:
- Content: Adequate coverage of key elements but lacking some depth
- Data Story: Clear narrative but lacking some engagement or coherence
- Visual Support: Adequate visual support with minor issues
- Delivery: Competent delivery with some room for improvement
- Q&A Response: Adequate responses to most questions
For a D-F grade:
- Content: Incomplete or superficial treatment of the project
- Data Story: Disjointed or unclear narrative
- Visual Support: Poor or distracting visual elements
- Delivery: Unclear, disorganized, or difficult to follow
- Q&A Response: Unable to address questions effectively
