Task

You need to apply what you have learned and prepare a convincing introduction for your master’s thesis.

During the course of the semester, you will go through the key stages of academic writing. These include finding a topic, writing a draft of the introduction to your master’s thesis, gathering feedback, revising your introduction and reflecting on the process.

You need to complete several deliverables as part of this process. These will be marked on an individual basis.

Deliverables

The portfolio exams consist of the following components:

The initial version of the introduction (2–3 pages) to your master’s thesis (template)
The AI review agent (system prompt) and rationale documenting how it operationalizes course content (template)
The reflection and revision of your introduction, including the revised version (template)

For dates and deadlines, please see the schedule

Grading

The deliverables are graded individually and determine the grade with the following weights:

20 points for the initial version of the introduction
30 points for the AI review agent (prompt and rationale)
50 points for the reflection and revision

Note: Compliance with formal requirements (see templates), good scientific practice, and submission of the roundtable protocol (date, participants, feedback given, feedback taken) are critical pass criteria. Non-compliance results in automatic failure (e.g., plagiarism, even light forms; missing Roundtable Protocol).

Introduction

An introduction that fulfills expectations has the following characteristics:

Structure: implements the structure discussed in class
Substance: addresses a clear and significant problem; states and justifies the research question; references relevant literature and points to a potential contribution
Narrative: tells a conclusive story that is well aligned to its audience and testifies professionalism
Language and form: uses appropriate vocabulary, consistent style and tone, and conventional technical formatting

Review agent

A review agent that fulfills expectations has the following characteristics:

Prompt design: the system prompt is reproducible, operationalizes specific course concepts (e.g., CARS, peer-review heuristics, structural principles from Macgilchrist (2014)), covers the relevant dimensions of an introduction, and specifies the agent’s output behavior and constraints
Rationale: design choices are justified and explicitly mapped to course content; prompt iteration (what was tried, failed, adjusted) is documented
Critical reflection: the agent’s limitations are critically evaluated, and where human feedback remains essential is identified

Reflection

A reflection that fulfills expectations has the following characteristics:

Synthesis: peer roundtable feedback and AI Review Agent feedback are summarized; recurring themes and tensions are identified
Comparative analysis: differences between AI and peer feedback are surfaced; decisions about what to adopt and what to reject are explained with reasons
Implementation and revision quality: concrete revisions are connected to specific feedback with examples; the revised introduction shows notable overall improvement and meets the criteria of a good academic introduction (structure, substance, narrative, language)
Meta-reflection: learning about the writing process, working with AI feedback, and giving feedback to peers is reflected upon

A note on grades

It is unlikely that every student will receive a very good grade, i.e. deliver an outstanding performance - see the meaning of grades. Instead, it is to be expected that the grades will spread across the spectrum.

Grade	Meaning
`1` — very good	A truly outstanding achievement that (not only) shows no deficiencies in the criteria mentioned, but also gives both the supervisor and external assessors an excellent impression.
`2` — good	Work that exceeds the average requirements/performance and is easily recognizable and presentable to the outside world as a “good performance”.
Note	2.5 is the average of passed assessments, i.e., an “average performance”
`3` — satisfactory	A performance that achieves the desired goal “to a satisfactory extent”; however, deficiencies can be identified here and there.
`4` — sufficient	A performance that “still adequately satisfies” the requirements, but deviates from the expectations placed on it in several ways.
`5` — not sufficient	A performance that does not meet several of the criteria mentioned.

Literature

Macgilchrist, F. (2014). Academic writing. UTB GmbH.