Syllabus

Course Overview

Course Description

Review of advanced quantitative procedures commonly applied in public administration research, with emphasis on multivariate models found in leading journals in the discipline. Illustrative topics include specialized regression models, generalized linear models, event history models, mixed and multilevel models, and structural equation models applied to public administration.

Course Objectives

This course explores advanced quantitative methods in applied social science research, with particular attention to public policy and management. It emphasizes how methods work, their underlying assumptions, and their intended purposes. Hands-on work provides practice with contemporary analytical tools. Students will build a methodological foundation, learn to judge which methods are appropriate in different contexts, implement basic models independently, and produce an original empirical paper emphasizing the data, methods, and results sections.

Learning Outcomes

Upon completion of this course, students will be able to:

Explain foundational concepts in statistical learning, including the distinction between prediction and inference, supervised and unsupervised approaches, and the bias-variance tradeoff.
Estimate and interpret linear and logistic regression models, including regularized approaches such as ridge regression and the lasso.
Implement supervised learning methods for classification and regression, including tree-based methods, random forests, boosting, and support vector machines.
Apply unsupervised learning methods such as principal components analysis and clustering to identify structure in data.
Communicate results of a statistical learning analysis in a clear, well-structured manuscript and presentation, with effective tables and visualizations.

Course Structure

This hybrid course includes asynchronous and synchronous components. Asynchronous components are delivered through the course website. Learning activities include readings, videos, presentations, quizzes, peer evaluations, and individual projects.

Course Prerequisites

PA 715 Quantitative Policy Analysis and PA 765 Quantitative Research in Public Administration, or equivalent coursework. Students should enter the course able to:

Interpret coefficients, standard errors, confidence intervals, and hypothesis tests.
Use the language of quantitative research, including variables, parameters, models, estimators, and statistical significance.
Conduct and interpret t-tests, chi-square tests, ANOVA, and ordinary least squares regression.
Use statistical software to manage data, run analyses, and generate output.
Read empirical research with statistical content and interpret tables, figures, and regression results.

Textbooks

Required Texts

Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani, and Jonathan Taylor, An Introduction to Statistical Learning: With Applications in Python (opens in a new tab). Springer, 2023. This course refers to this book as ISLP.

Optional Texts

J. Scott Long, Regression Models for Categorical and Limited Dependent Variables (opens in a new tab). SAGE Publications, 1997.
Frank E. Harrell Jr., Regression Modeling Strategies (opens in a new tab) (2016).
Peter K. Dunn and Gordon K. Smyth, Generalized Linear Models With Examples in R (opens in a new tab) (2018).
Sebastian Raschka and Vahid Mirjalili, Python Machine Learning (opens in a new tab) (2019).
Peter Martin, Regression Models for Categorical and Count Data (opens in a new tab) (2021).
Hal Daumé III, A Course in Machine Learning (opens in a new tab).

Tools

The primary programming environments are Python and R, used through Google Colab (opens in a new tab) during class meetings. Students may use other platforms if they can complete the assignments. Stata will be used for selected survival-analysis and advanced difference-in-differences exercises.

Digital Course Components

Course Website: Course materials, readings, and publicly available videos.
Moodle: Grades, announcements, and important updates.
Google Forms: Assignment submissions and peer reviews.
Panopto: Recordings available only to enrolled students.
Google Colab: Browser-based Python and R environment for exercises, analysis, and replication.

Grading & Feedback

Grading Components & Weighting

Student performance will be evaluated using the following components:

Course grading components and percentage of final grade
Component	Percentage of Final Grade
Problem Sets	20%
Project	55%
Engagement	25%

Participation credit is earned through active engagement in class discussions and in-class activities. Attendance itself is not recorded as part of the course grade.

Final Grade Breakdown

Final course scores correspond to the following letter grades:

Final course score ranges and corresponding letter grades
Grade	Final Course Score	Grade	Final Course Score
A	94–100	A-	90–93.9
B+	85–89.9	B	83–84.9
B-	80–82.9	C+	75–79.9
C	73–74.9	C-	70–72.9
D+	68–69.9	D	63–67.9
D-	60–62.9	F	0–59.9

Assignment Submission & Feedback

Submit assignments through the designated links under the corresponding assignment tabs. Assignments submitted by email will not be graded. Grades and feedback will be returned through Canvas.

Late Assignments

Late assignments will be accepted with a 25% deduction for each day submitted late.

Credit-Only (S/U) Grading

For students taking the course for credit only, work equivalent to C- or better earns an S (Satisfactory); work below C- earns a U (Unsatisfactory). See the Credit-Only Courses regulation (opens in a new tab).

Auditing

Auditing is approved on a case-by-case basis. Contact the instructor for approval and consult the Audit regulation (opens in a new tab) for requirements and forms.

Course Policies & Procedures

Inclusion Statement

It is my intent that students from all diverse backgrounds and perspectives be well-served by this course. I aim to create an inclusive environment where students’ learning needs are met, and the diversity of experiences, identities, and viewpoints enriches our shared learning. I strive to present materials and activities that acknowledge and respect differences in gender identity, sexual orientation, disability, age, socioeconomic status, ethnicity, race, religion, culture, and perspective. Your feedback on enhancing the inclusivity and effectiveness of the course is welcome. If you have suggestions for improving your experience or that of your peers, please feel free to reach out. I also have made an effort to avoid scheduling major deadlines during significant religious holidays. If any deadline conflicts with your religious observances, please contact me at serena_kim@ncsu.edu so we can make appropriate adjustments.

Student Rules of Conduct

Students and faculty share responsibility for maintaining an appropriate and respectful learning environment. NC State REG 11.35.05 Code of Student Conduct (opens in a new tab) sets expectations for behavior in both virtual and physical classrooms, as well as consequences for violations. While diverse viewpoints and interpretations of course content are welcome, any behavior that disrupts others’ ability to learn and succeed will be addressed. Students are expected to adhere to the following rules of conduct to maintain a productive and respectful learning environment:

Respect and Inclusion: Treat all members of the class—peers, instructors, and guests—with respect. Discrimination, harassment, or inappropriate behavior of any kind is strictly prohibited. This includes professional courtesy and sensitivity toward individuals and topics involving race, color, national origin, gender identity, sexual orientation, disability, age, socioeconomic status, ethnicity, religion, culture, perspective, or other background characteristics.
Engage Constructively: Contribute to class discussions and group work in a positive and respectful manner. Allow others the opportunity to share their perspectives without interruption or judgment.
Maintain Academic Integrity: Follow NC State’s policies on academic integrity. Plagiarism, cheating, or unauthorized collaboration on assignments is not allowed.
Be Prepared and Focused: Complete all assigned readings, tasks, and exercises before class. During sessions, silence personal devices, avoid distractions, and stay engaged. Activities such as phone calls, use of headphones, persistent talking, whispering, and web surfing unrelated to the course are prohibited.
Communicate Professionally: Use respectful, professional language in all communications, including emails, discussions, and written assignments.
Respect Class Time and Privacy: Arrive on time for all meetings and inform the instructor in advance if you need to arrive late or leave early. Do not record or share course content, discussions, or other students’ work without explicit permission from the instructor and all involved parties.
Use Course Tools Appropriately: Use course-related digital platforms solely for their intended educational purposes.

Failure to adhere to these behavioral standards may result in disciplinary action. Significant violations may lead to a failing grade for the course and will be reported to the appropriate authorities.

Incomplete Grades and Withdrawals

Information on incomplete grades can be found in REG 02.50.03 – Grades and Grade Point Average (opens in a new tab). If you encounter a serious disruption to your work not caused by you and you would have otherwise successfully completed the course, contact your instructor as soon as you can to discuss the possibility of earning an incomplete in the course for the semester, including an agreement on when the remaining work must be done in order to change the grade to the appropriate letter grade. If you must withdraw from a course or from the university due to hardship beyond your control, see the Withdrawal Process and Timeline (opens in a new tab) for information and instructions.

Artificial Intelligence (AI) Policy

Using AI as a Learning Aid

Generative and agentic artificial intelligence (AI) tools are optional learning aids in this course. These tools include chatbots, code assistants, spreadsheet copilots, automatic-analysis tools, and generative features embedded in other software. Students are not required to use an AI assistant, will not be disadvantaged for choosing not to use one, and can complete every assignment without one.

When an assignment requires students to run, evaluate, or adapt AI models as part of the course content, that required activity is distinct from using an AI assistant to produce the student’s work. Conventional calculators, statistical software, and execution of student-written code are also not considered AI assistance. Assignment-specific instructions may impose stricter conditions than this policy.

Core principle: AI may support learning and execution, but it may not replace the statistical reasoning, research judgment, source evaluation, or communication that an assignment is designed to assess.

Permitted Assistance

Concept Learning: AI may explain concepts, terminology, error messages, or course-related ideas and provide additional examples.
Analytical Brainstorming: AI may suggest research questions, variables, models, statistical tests, diagnostics, alternative interpretations, or study limitations. Students must evaluate these suggestions and make the final analytical decisions.
Code Explanation and Debugging: After making a genuine attempt, students may ask AI to explain or help debug their existing code. A genuine attempt should show the student’s intended logic, even if the code is incomplete or contains errors.
Editing Mechanics: AI may identify and correct grammar, spelling, punctuation, and typographical errors in text the student has already written. Assistance must remain limited to correcting errors rather than rewriting, expanding, or reorganizing the content.
Formatting: AI may help format student-created tables or organize verified references in a consistent citation style, provided it does not add, remove, invent, or substantively change information.

Prohibited Assistance

Generating Code From Scratch: Students may not submit an assignment prompt to AI and ask it to produce the necessary code, answer, or complete solution. AI may not replace the student’s initial coding attempt.
Drafting or Rewriting: AI may not draft, expand, substantially rewrite, or reorganize reports, documentation, analytical explanations, presentation content, posters, or speaker notes.
Substituting for Analysis: Students may not rely on AI to analyze an uploaded dataset or provide final calculations, results, interpretations, conclusions, or policy recommendations that they have not independently produced, verified, and can explain.
Generating Final Figures: AI may not create a final data figure as an image or generate its plotting code from scratch. AI may help debug student-written visualization code, but the submitted figure must be traceable to the underlying data and reproducible.
Fabricating Evidence: AI may not be used to generate or alter citations, sources, quotations, data, observations, survey responses, participants, or research findings. AI output is not an authoritative source.
Independent Assessments: AI may not be used on quizzes or any other assessment designated for independent work.

Data, Privacy, and Course Materials

Do not enter identifiable, confidential, respondent-level, or otherwise restricted data into an AI tool unless the instructor explicitly authorizes both the tool and the proposed data use. Raw survey responses, another student’s work, and restricted course materials may not be uploaded to public or personal AI accounts. When university data are involved, students must use an appropriately authorized account and follow NC State’s approved-tool and data-classification requirements (opens in a new tab).

Responsibility and Evidence of Learning

Students are responsible for every element they submit, regardless of whether AI provided assistance. AI recommendations are suggestions, not answers. Students must verify factual claims using authoritative sources and reproduce numerical results in Python, R, Excel, Stata, or another approved analytical program. Submitted figures must be reproducible through the underlying data and the student’s code or formulas.

Students must be prepared to explain their code, formulas, analytical choices, results, writing, and visualizations; describe what they attempted before consulting AI; identify and modify AI-assisted portions of their work; and answer brief questions or demonstrate part of their workflow when requested. For group work, every member must understand and be able to explain the group’s use of AI.

Demonstration of Learning: The instructor may ask any student to briefly explain, reproduce, or modify part of submitted work—such as code, formulas, analytical decisions, results, figures, or writing—particularly when questions arise about how the work was produced or whether AI was used within course rules. This may take the form of follow-up questions or a short live walkthrough. An inability to demonstrate reasonable understanding may affect whether the work satisfies the assignment’s learning objectives and may prompt further review under the university’s academic-integrity procedures.

Submitted notebooks, scripts, spreadsheet formulas, drafts, version history, presentations, and brief oral explanations may be considered alongside any demonstration when assessing understanding and authorship.

AI-Use Statement

Unless an assignment explicitly provides an exemption, include a brief AI-use statement identifying: (1) the tool used; (2) the purpose for which it was used; (3) the work completed before consulting the tool; (4) the advice or corrections incorporated; and (5) how the output was verified. Students should retain relevant AI conversations and earlier versions of their work in case clarification is needed, but a complete transcript of every prompt is not required.

Example: “I used Gemini to help diagnose an error in my logistic-regression code after attempting the model myself. It suggested that my outcome variable had been stored as text. I inspected the variable type, corrected it in Python, reran the model, and verified the output against the course example.”

Routine built-in spelling or grammar flags that do not rewrite sentences and non-generative formatting features do not require disclosure. If no generative or agentic AI assistant was used, state: “No generative or agentic AI assistant was used in completing this assignment.”

Academic Integrity

Using AI outside the boundaries established by this policy or an assignment’s instructions may constitute unauthorized assistance, falsification, misuse of academic materials, or plagiarism under NC State’s academic-integrity policies. Suspected misuse will be addressed through the university’s established academic-integrity and student-discipline procedures.

AI-detection or similarity reports may prompt a conversation, but they will not, by themselves, establish academic misconduct. Human review, process evidence, and the student’s ability to explain the submitted work will be considered.

University Policies

Academic Integrity and Honesty

Students are required to comply with the university policy on academic integrity found in the Code of Student Conduct 11.35.01, sections 8 and 9 (opens in a new tab). Therefore, students are required to uphold the Pack Pledge: “I have neither given nor received unauthorized aid on this test or assignment.” Violations of academic integrity will be handled in accordance with the Student Discipline Procedures (opens in a new tab). Please refer to the Academic Integrity overview (opens in a new tab) for a detailed explanation of the university’s policies and common understandings related to them.

Student Privacy

Originality-checking software: Software such as Turnitin may be used in this course to detect the originality of student submissions.

Class recording statement: In-class sessions are recorded in such a way that might also record students in this course. These recordings will not be used beyond the current semester or in any other setting outside the course.

Class privacy statement: This course requires online exchanges among students and the instructor, but not with persons outside the course. Students may be required to disclose personally identifiable information to other students in the course through electronic tools such as email or web postings where relevant to the course. Examples include online discussions of class topics and posting student coursework. All students are expected to respect each other’s privacy by not sharing or using such information outside the course.

Other Policies

Students are responsible for reviewing the NC State University Policies, Rules, and Regulations (PRRs) that pertain to their course rights and responsibilities:

Course Evaluations

ClassEval is the end-of-semester survey for students to evaluate instruction in all university classes. The survey is administered online and includes 12 closed-ended questions and three open-ended questions. Deans, department heads, and instructors may add a limited number of their own questions to these 15 common-core questions.

Each semester, students’ responses are compiled into a ClassEval report for every instructor and class. Instructors use the evaluations to improve instruction and include them in their promotion and tenure dossiers, while department heads use them in annual reviews. The reports are included in instructors’ personnel files and are considered confidential.

Online class evaluations are available during the last two weeks of the semester for full-semester courses and the last week of shorter sessions. Students will receive an email directing them to a website to complete class evaluations. Evaluations become unavailable at 8:00 a.m. on the first day of finals. See more information about ClassEval (opens in a new tab).

ClassEval website (opens in a new tab)
ClassEval Help Desk: classeval@ncsu.edu

Student Resources

Academic and Student Support

Academic and Student Affairs (opens in a new tab) maintains a website with links for student support on campus, including academic support, community support, health and wellness, financial hardship or insecurity, and more. Visit Find Help on Campus (opens in a new tab).

Disability Resources

Reasonable accommodations will be made for students with verifiable disabilities. To take advantage of available accommodations, students must register with the Disability Resource Office (DRO) (opens in a new tab). For more information on NC State’s policy on working with students with disabilities, see the DRO Policies, Rules and Regulations (opens in a new tab) and REG 02.20.01 – Academic Accommodations for Students with Disabilities (opens in a new tab). Please contact the instructor at serena_kim@ncsu.edu to submit an accommodation letter within the first three weeks of the semester.

Safe at State

At NC State, we take the health and safety of students, faculty, and staff seriously. The Office of Equal Opportunity (opens in a new tab) supports the university community by providing services and resources to guide individuals in obtaining the help they need. See the Safe at State webpage (opens in a new tab) for resources.

Supporting Fellow Students in Distress

As members of the NC State Wolfpack community, we each share a personal responsibility to express concern for one another and to ensure that this classroom and the campus as a whole remain healthy and safe environments for learning. Occasionally, you may come across a fellow classmate whose personal behavior concerns or worries you, either for the classmate’s well-being or your own. If you feel this way, please report the behavior through the NC State CARES website (opens in a new tab). Although you can report anonymously, it is preferred that you share your contact information so the team can follow up with you personally.