Problem & Significance
Precision, tractability, and importance of the research problem, including a clear explanation of why it matters to the relevant scholarly or applied audience.
Project
Weight
55% of final grade
Components
3 presentations + final paper
Final Deliverable
Conference-ready paper
Work Format
Individual
You will develop one original empirical research project across the semester. Choose an applied research question and dataset that support meaningful quantitative analysis; the topic does not need to be limited to public administration. The project culminates in a conference-ready empirical research paper.
You may use any appropriate quantitative analysis technique, including methods learned in other courses. Depending on your question and data, possibilities include linear regression, generalized linear models, multinomial or ordinal models, count models, multilevel or panel models, time-series or spatial analysis, causal-inference designs, survival analysis, machine learning, and quantitative text analysis. This menu is illustrative rather than exhaustive; consult the instructor with questions about the fit or feasibility of a proposed approach.
Three project touchpoints provide structured opportunities to present your progress and completed study, and to receive feedback across the semester. The Project grade covers your presentation in each of the three workshops and the final research paper. Serving as a discussant and contributing feedback to classmates are evaluated separately under Engagement.
Checkpoint materials support the workshop presentations and discussions; they are not separate papers. Detailed submission instructions will be posted with each component.
Touchpoint 1
Proposal/Pitch
Introduce the research problem and present a focused plan for the empirical project. Submit either a 1–2 page proposal or a brief slide deck one week before Workshop 1. Slides are optional for the Workshop 1 presentation; you may present with or without them. Give a concise presentation followed by discussant-led questions and class feedback.
What to include
- State the research question and explain why it matters.
- Describe the proposed argument, expectations, or hypotheses.
- Identify the anticipated data, unit of analysis, and general analytical approach.
- Flag the most important uncertainty or decision on which feedback would be useful.
- Optional: Briefly connect the project to relevant literature and identify its prospective scholarly or applied contribution.
How the Proposal/Pitch is evaluated
The four criteria below total 100% of Touchpoint 1.
Research Question & Alignment
Clarity and answerability of the research question and its alignment with the project’s intended outcome, population, scope, and analytical goal.
Argument & Expectations
Coherence of the proposed argument and the clarity and internal alignment of any hypotheses or empirical expectations appropriate to the project.
Preliminary Design, Feasibility & Pitch
Plausibility of the proposed data and analytical approach, recognition of consequential uncertainties, and disciplined communication in the submitted proposal or slide deck and the presentation.
View the detailed Proposal/Pitch rubric Expand to compare doctoral-level performance descriptions for all four criteria.
The performance band determines the percentage earned within each criterion; the Weight column determines that criterion’s contribution to Touchpoint 1.
| Criterion | Exemplary 95–100% |
Proficient 80–94.9% |
Developing 40–79.9% |
Needs Improvement Below 40% |
Weight |
|---|---|---|---|---|---|
| Problem & Significance | Defines a consequential, tractable, and well-bounded problem and explains persuasively why it matters to the relevant scholarly or applied audience. | Defines a clear and relevant problem whose value is evident. The scope, stakes, intended audience, or explanation of significance requires limited refinement. | Identifies a general topic, but the research problem is broad or weakly bounded and its significance is asserted more than demonstrated. | Does not establish a coherent, tractable research problem or explain why the proposed study matters. | 30% |
| Research Question & Alignment | States a precise, answerable research question that is tightly aligned with the intended outcome, population, scope, and analytical goal. | Presents a clear and answerable question with generally appropriate alignment. Minor ambiguity remains in scope, constructs, population, outcome, or analytical goal. | The question is overly broad, combines multiple studies, or is only partly answerable, with consequential misalignment among the question, scope, or intended analysis. | Lacks an answerable research question or presents a question that cannot guide a feasible empirical study. | 25% |
| Argument & Expectations | Presents a coherent, logically developed argument. Hypotheses or expectations, when appropriate, are explicit, testable, and aligned with the proposed study and observable evidence. | Presents a plausible argument with generally aligned expectations. Minor ambiguity remains in the reasoning, constructs, direction, or connection to observable evidence. | The argument is underdeveloped, and hypotheses or expectations are vague, weakly derived, or inconsistently aligned with the proposed study. | Lacks a coherent argument or any usable expectations connecting the proposed study to observable evidence. | 25% |
| Preliminary Design, Feasibility & Pitch | Proposes obtainable data, a defensible unit and scope, and an analytical strategy suited to the question. Identifies consequential risks and feedback needs, and communicates the plan with precision and economy. | Presents a workable preliminary design and prepared pitch. Remaining issues in access, measurement, method fit, scope, or prioritization appear resolvable within the semester. | The design remains speculative or overextended; data access, unit of analysis, method fit, or feasibility is uncertain. The materials or presentation do not clearly prioritize decisions requiring feedback. | Offers no plausible empirical path, omits essential checkpoint materials, or is insufficiently prepared to explain and defend the proposed project. | 20% |
Touchpoint 2
Data & Methods Check-in
Present the project’s operational research design before completing the main analysis. Submit either a 2–4 page data and methods memo or a brief slide deck one week before Workshop 2. Slides are optional for the Workshop 2 presentation; you may present with or without them. Show the data you are actually using, not only a description of data you plan to obtain.
What to include
- Document the data source, sample, unit of analysis, scope, and timeframe.
- Explain how the central concepts are measured and identify key variables.
- Describe the proposed analytical strategy and why it fits the question.
- Include an initial descriptive table, figure, or other concise preview of the data.
- Identify feasibility concerns, limitations, or methodological decisions that remain unresolved.
How the Data & Methods Check-in is evaluated
The four criteria below total 100% of Touchpoint 2.
Data Provenance, Sample & Reproducibility
Transparency about data origin, construction, population and sample, unit of analysis, scope, missingness, cleaning decisions, access constraints, and reproducible preparation.
Measurement & Descriptive Evidence
Validity of operational definitions, coding and transformations, alignment between constructs and measures, and use of descriptive evidence to diagnose the realized data.
Model Specification & Identification Logic
Fit between the research question, inferential or predictive target, outcome structure, dependence in the data, estimator, assumptions, uncertainty, and strength of the claims sought.
Diagnostics, Limitations & Research Readiness
Recognition of threats and model risks, planned diagnostics and robustness checks, calibrated limitations, concrete next steps, and readiness to complete a defensible analysis.
View the detailed Data & Methods rubric Expand to compare doctoral-level performance descriptions for all four criteria.
The performance band determines the percentage earned within each criterion; the Weight column determines that criterion’s contribution to Touchpoint 2.
| Criterion | Exemplary 95–100% |
Proficient 80–94.9% |
Developing 40–79.9% |
Needs Improvement Below 40% |
Weight |
|---|---|---|---|---|---|
| Data Provenance, Sample & Reproducibility | Documents the realized dataset with sufficient precision to reconstruct it: provenance and version, acquisition, population and sample, inclusion rules, unit, coverage, cleaning, missingness, access or ethical constraints, and reproducible preparation. | Documents the dataset, sample, scope, and preparation clearly overall. Minor omissions in provenance, exclusions, missing-data handling, versioning, or reproducibility do not obscure what the observations represent. | Uses real data but leaves consequential uncertainty about provenance, selection, unit, scope, cleaning, missingness, or reproducibility. It is difficult to determine what population the analysis can represent. | Does not present usable data or omits essential information needed to understand, evaluate, or reproduce the analytical sample. | 25% |
| Measurement & Descriptive Evidence | Operationalizes central constructs defensibly, documents coding and transformations, evaluates validity and measurement limitations, and uses well-chosen descriptive evidence to reveal distributions, missingness, outliers, and consequential structure. | Defines and codes the main measures appropriately and provides useful descriptive evidence. Some validation, transformation rationale, or diagnosis of data quality requires further development. | Measures are only partly aligned with the constructs or insufficiently documented. Descriptive evidence is thin, mislabeled, or not used to identify problems that could affect modeling. | Key constructs lack defensible measures, coding is opaque or erroneous, or no meaningful descriptive examination of the data is provided. | 25% |
| Model Specification & Identification Logic | States the inferential, descriptive, causal, or predictive target and selects a model consistent with the outcome, sampling process, dependence, functional form, and available information. Assumptions, identification logic, and uncertainty estimation match the claims. | Chooses a defensible analytical strategy that generally fits the question and data. The estimand, dependence structure, functional form, assumptions, or uncertainty plan needs limited refinement. | The method is named but weakly justified, or important features of the outcome and data structure are ignored. The proposed claims exceed what the design or identification logic can support. | The proposed model is inappropriate, technically incoherent, or disconnected from the research question and realized data. | 30% |
| Diagnostics, Limitations & Research Readiness | Prioritizes credible threats, specifies diagnostics and robustness or sensitivity analyses, states limitations without undermining the study’s purpose, and presents a feasible path from current data to a completed, defensible analysis. | Recognizes the main risks and proposes useful diagnostics and next steps. Some robustness strategy, prioritization, or articulation of limitations remains incomplete but the project is ready to proceed. | Risks and limitations are discussed generically, diagnostics are vague, or major feasibility and analytical decisions remain unresolved without a concrete plan. | Shows little awareness of consequential assumptions or threats and lacks a feasible plan for completing and validating the analysis. | 20% |
Touchpoint 3
Final Paper & Presentation
Touchpoint 3 includes two separately evaluated deliverables: the conference-ready research paper and a 15-minute presentation of the completed study.
Paper expectations
Each student will prepare an original research paper that presents a complete empirical research design and analysis. The paper should include the following sections, or an equivalent structure appropriate to the project:
- Abstract
- Introduction
- Brief literature review
- Data
- Methods
- Results
- Discussion and conclusion
The paper may not exceed 7,000 words, excluding the abstract, references, and appendices. Include at least one visualization of the analysis results.
Presentation expectations
Deliver a 15-minute, conference-style presentation and participate in a substantive discussant-led Q&A. Circulate the slide deck one week before Workshop 3. The suggested sequence below may be adapted when the project requires a different scholarly narrative.
- Title
- Motivation: 1–3 slides, including the research question.
- Literature: 1–3 slides positioning the project’s contribution.
- Estimation Methods: 1–3 slides explaining the data, measures, design, and analytical approach.
- Findings/Results: As many slides as needed to communicate the central evidence efficiently; include at least one figure when results are available.
- Discussion/Future Steps: 1–3 slides covering interpretation, limitations, implications, and next steps.
- Acknowledgements/References
All students are expected to ask questions of one another. The presenter’s command of the study and responses during Q&A are evaluated here; classmates’ preparation and questions are evaluated separately under Engagement.
How the Final Paper & Presentation are evaluated
The Paper and Presentation are evaluated separately and together constitute Touchpoint 3.
Paper evaluation
The four criteria below total 100% of the Paper portion.
Contribution & Scholarly Framing
Importance and originality of the question, synthesis of relevant literature, conceptual precision, and clarity about how the study advances scholarly or applied understanding.
Research Design, Data & Measurement
Transparency and defensibility of the design, data provenance and sample, operationalization, estimand or predictive target, identification logic, and scope of warranted inference.
Analysis, Diagnostics & Reproducibility
Technical correctness of estimation, treatment of uncertainty and dependence, diagnostic and robustness evidence, reproducible workflow, and accurate tables and figures.
Interpretation, Limitations & Conference-Ready Writing
Substantive interpretation, calibration of claims to evidence, meaningful limitations and implications, coherent organization, precise prose, complete references, and professional presentation.
View the detailed Paper rubric Expand to compare doctoral-level performance descriptions for all four criteria.
The performance band determines the percentage earned within each criterion; the Weight column determines that criterion’s contribution to the Paper portion.
| Criterion | Exemplary 95–100% |
Proficient 80–94.9% |
Developing 40–79.9% |
Needs Improvement Below 40% |
Weight |
|---|---|---|---|---|---|
| Contribution & Scholarly Framing | Advances a consequential, precisely bounded question; synthesizes the relevant literature into a compelling gap or tension; and delivers a clear theoretical, methodological, or substantive contribution. | Establishes a meaningful question and credible contribution with strong literature coverage. Some novelty, theoretical mechanism, scope, or positioning could be sharpened for conference submission. | The question is identifiable but broad or derivative. Literature is mainly descriptive, and the claimed contribution is weakly distinguished from prior work. | Lacks a coherent research question, adequate engagement with relevant scholarship, or a discernible contribution. | 20% |
| Research Design, Data & Measurement | Aligns the inferential or predictive target, design, data-generating process, sample, measures, and method. Documents provenance and analytic choices transparently and calibrates identification and generalization claims to the evidence. | Presents a coherent, defensible design with well-documented data and measures. Minor gaps remain in identification, validity, sampling, missingness, ethics, or scope of inference. | Important design choices are underjustified or incompletely documented. Measurement, selection, identification, or generalizability concerns materially weaken the claims. | The design and data cannot support the stated question, or essential information about the sample, measures, or inferential logic is absent. | 25% |
| Analysis, Diagnostics & Reproducibility | Implements the analysis correctly and transparently, estimates uncertainty appropriately, addresses dependence and assumptions, reports informative diagnostics and robustness checks, and produces reproducible, publication-quality evidence. | The analysis is correct and reproducible overall, with appropriate uncertainty and clear evidence. Additional diagnostics, sensitivity tests, specification justification, or visual refinement would strengthen confidence. | Core analysis is present but includes consequential technical, diagnostic, reproducibility, or reporting gaps. Results may be sensitive to unexamined assumptions or specifications. | Major analytical errors, opaque workflow, absent uncertainty, or unsupported output prevent credible evaluation of the findings. | 30% |
| Interpretation, Limitations & Conference-Ready Writing | Distinguishes statistical from substantive meaning, answers the question with calibrated claims, treats limitations and implications analytically, and presents a cohesive, polished, properly cited manuscript ready for a conference audience. | Interprets the evidence accurately and presents a well-organized scholarly paper. Claims, limitations, implications, prose, or references require limited revision before conference submission. | Interpretation is partly mechanical or overextended; limitations and implications are generic; or organization, prose, visual integration, and citation practices require substantial revision. | Misinterprets central findings, makes claims unsupported by the design, or presents writing and organization too incomplete for scholarly evaluation. | 25% |
Presentation evaluation
The four criteria below total 100% of the Presentation portion.
Scholarly Narrative & Structure
Coherent progression from motivation and question through literature, methods, findings, and discussion, with disciplined prioritization appropriate to a 15-minute conference presentation.
Methods & Evidence Communication
Accurate, audience-aware explanation of the design and estimation strategy, transparent communication of uncertainty, and effective selection and interpretation of the central results.
Visual Design & Delivery
Readable, purposeful slides and figures; professional pacing and delivery; accessible technical language; and effective use of time without reading from a script.
Command of Project & Q&A
Ability to defend analytical choices, distinguish established findings from uncertainty, respond directly and thoughtfully, and use questions to clarify limitations and future steps.
View the detailed Presentation rubric Expand to compare doctoral-level performance descriptions for all four criteria.
The performance band determines the percentage earned within each criterion; the Weight column determines that criterion’s contribution to the Presentation portion.
| Criterion | Exemplary 95–100% |
Proficient 80–94.9% |
Developing 40–79.9% |
Needs Improvement Below 40% |
Weight |
|---|---|---|---|---|---|
| Scholarly Narrative & Structure | Builds a compelling and economical research narrative, foregrounds the contribution, and allocates time strategically across motivation, literature, methods, findings, and discussion within the 15-minute limit. | Follows a clear scholarly structure and communicates the main contribution within the allotted time. Minor imbalance, repetition, or transitions reduce efficiency but not comprehension. | The sequence is recognizable but important elements are omitted, overdeveloped, or weakly connected. Timing prevents adequate treatment of the central evidence or contribution. | Lacks a coherent structure, substantially violates the time limit, or does not communicate the study’s question, approach, findings, and meaning. | 25% |
| Methods & Evidence Communication | Explains the design and estimation choices with technical accuracy and audience awareness, communicates uncertainty and substantive magnitude, and uses selected results to support appropriately bounded conclusions. | Accurately explains the main methods and findings. Some assumptions, uncertainty, estimand, substantive interpretation, or connection between evidence and claims needs clarification. | Methods are oversimplified or difficult to evaluate, and findings are presented mechanically, selectively, or without adequate uncertainty and substantive interpretation. | Misstates the analytical approach or results, omits the central evidence, or draws conclusions the presented analysis cannot support. | 30% |
| Visual Design & Delivery | Uses legible, uncluttered slides and at least one effective results figure; defines technical terms; maintains professional pacing and presence; and speaks to the audience rather than reading from slides or notes. | Slides, figures, and delivery are clear and professional overall. Minor issues in density, labeling, accessibility, pacing, vocal delivery, or reliance on notes do not obscure the argument. | Dense or poorly labeled visuals, uneven pacing, unexplained jargon, or heavy reliance on notes makes parts of the presentation difficult to follow. | Visuals are missing, misleading, or unreadable, or delivery is unprepared and prevents the audience from understanding the project. | 20% |
| Command of Project & Q&A | Responds directly and analytically, defends choices with evidence, acknowledges uncertainty without evasion, and uses questions to articulate limitations, alternative explanations, robustness needs, and productive next steps. | Answers questions accurately and demonstrates strong command of the project. Some responses could be more direct, technically precise, or reflective about limitations and alternatives. | Responses show partial command, repeat prepared material, or struggle to explain key design choices, assumptions, findings, or limitations. | Cannot explain or defend essential elements of the research or responds in ways that materially misrepresent the design or findings. | 25% |