Students are required to summarize and discuss the available scientific evidence on a chosen public-sector policy question. The review should i) identify the public-sector policy problem and the corresponding policy, ii) define the scope of the review, describe the literature search and (if applicable) how you used AI tools critically for assistance iii) summarize the available evidence and how this evidence was produced by the study authors, iv) and critically discuss important gaps in existing research. This exercise offers students an opportunity to build their expertise in a particular policy area and begin to develop their dissertation topic.

When choosing a policy question, students should focus on ones that are amenable to evidence rather than normative questions. For instance, instead of asking whether a policy should be implemented, consider if a policy achieves a particular goal. Given the constraints of time and word count, it’s essential to select tractable policy questions. For example, asking “What works to boost naturalization rates?” might be too broad, while “Does New York fee reduction increase naturalization rates?” could be overly specific. A balanced question might be “Do fee reduction policies boost naturalization rates?”

Requirements

The review must be concise, not exceeding 3,000 words. It’s essential to choose a public policy question that allows for a review of at least five studies (absolute minimum). The majority of studies should be published in peer-reviewed journals in Economics and/or Political Science.

There is no complete list of acceptable journals. The purpose of this requirement is not to limit the type of work you can review, but to nudge you to focus on policy questions relevant to Political Science and Economics (broadly defined). If you find yourself in a situation in which the majority of your studies are published in, for example, Sociology journals or in general science journals (e.g., Nature Human Behavior, PNAS) please talk to me for feedback.

Adherence to grammar and spelling norms is expected. Students have to consistently employ either the Harvard or Chicago reference style. Please consult the UCL Library guide on referencing if you are not familar with referencing and consider using reference management software (UCL Library guide to reference management software). The final submission must be a PDF with a font size of 12, 1.5 line spacing, and page numbers.

Marking Criteria

  • Scope/number of studies and replicable literature search (about 15%)
  • Accurate, original and purposeful summary of the studies (about 20%)
  • Evidence of own critical and original analysis (about 30%)
  • Overall insight and originality (about 20%)
  • Purposeful writing, formatting, referencing (about 15%)

Political Science Department policy requires that penalty points be deducted for written works that are late and do not allow module convenors to grant extensions. Department policy also requires that penalty points be deducted for essays that exceed the maximum word limit. Any words in tables and graphs as well as footnotes must be included in the word count. The bibliography and (if applicable) appendices, abstract and acknowledgements are not included in the word count.

AI Guidelines

In this module, you are authorised and encouraged to seek assistance from AI tools. If you decide to use AI tools, you must disclose their use. It is always considered academic misconduct to copy and paste text generated by AI tools into your review (see next section). If you use AI tools to edit your writing (e.g., fixing typos or grammatical errors), disclosure of AI use in the first footnote—by mentioning the model name and number—is sufficient.

If you choose to use AI tools for any research purposes, you must critically engage with their output. Critical engagement with AI output depends on the specific use case, but it often requires a strategy to verify the output produced by the AI tool. This strategy must be described in your submission (e.g., in the main text as part of your methodology section). You must also name the AI tool(s) you used, include the model number(s), and list all prompts in the appendix.

For example, if you choose to use AI tools to assist with a literature search, you must describe a strategy to verify that the set of studies returned by the AI tool is complete and that no relevant studies are missing. In practice, this likely means complementing the AI output with another manual search strategy to ensure that other relevant studies are not omitted and that the studies deemed relevant by the AI tool are indeed relevant.

Academic Integrity

All research and writing for this module must be produced by the student. If you have another person do even parts of the research or the writing for you, this is considered academic misconduct. Any undisclosed use of AI tools is also considered academic misconduct. Any suspected Academic Misconduct will be investigated and can ultimately lead to module failure. For details see UCL Guidelines on academic misconduct.

Important Deadlines

  • Oct 31, 2 PM: Register policy question via Policy Question Register
  • Nov 14, seminar: Report on your progress in the seminar
  • Dec 12, 2 PM: Submit the complete bibliography Bibliography
  • Jan 13, 2 PM: Submit the complete review

Examples

The PUBL0097: Review Idea Repository provides a series of ideas for policy questions. You are welcome to—but not required—to pick a question from this repository or to use it for inspiration to define your own policy question.

The Moodle for this module includes example submissions from previous years for inspiration. The published examples below are meant to provide ideas and highlight the large scope of potential policy questions. Students are not expected to write a publication-ready literature review.

Further Readings