Reflection Essay

Introduction

In this assignment the students should critically reflect on the course to attain a better comprehension of its topics. Reflection is a powerful technique to reinforce their learning and professional development, and it helps them pay more attention to their practice.

The questions mentioned serve as a starting point to spark your critical analytical thought process. However, we recommend you also think of questions you personally would like to answer. Your reflection shouldn’t just be a descriptive retelling of what was covered in the course but should instead be analytical. Engage critically with the discussed topics. Write about what you were thinking and feeling during the course. What was good and bad about it? What sense can you make of the discussed situations? What else could have been done instead of what was presented? If you encounter such problems, what would you do?

Instructions

The core question for this reflection assignment is: “How will the concepts, introduced during the course, impact your next data science practice?”

You are required to write a reflection paper addressing this question. There are many ways to write such a document and there is no single perfect solution for this, so follow your intuition and your interests. As guidance, we encourage you to go through four phases of reflective activity. These phases are:

Phase I – Awareness of your current data science practice

What is your current data science practice (before attending the course)?
What is the context in which you practice data science (e.g., the environment you work in, the problems you face)?
What are the reasons for how you practice data science as you do?

Phase II – Clarify the new learning and how it relates to your current understanding

What was the course trying to achieve?
What is it that you have learned in this course that can improve your practice?
Where did the course propose interventions and why?
What are the consequences for yourself?
What are the consequences for society?

Phase III – Integrate the new learning into your current data science practice

How does this new learning relate to what you knew and did before?
How did the learning influence your decision-making during the course?
Has this experience changed your views on, for example, bias, fairness, or ethics?
How did the course affect your personal views?

Phase IV – Anticipate or imagine the nature of your future data science practice

How will you act in such a way that your practice is improved (as a result of the learning)?
What action choices do you have?
What are the consequences of these choices?
How do you feel about the improved practice?

Evaluation

Reflection Essays can be evaluated in many ways. Here are some criteria that can be used:

Name	Description	Measurement	The rule for failing	Comment
C1: Objective Task Fulfillment
1.1 Word count	The essay should contain a minimum of 1250 words.	Amount of written words excluding titles, keywords, and captions.	< 1240 words (10-word buffer)	Originally a maximum of 1500 words was the limit, which was now raised to 2500. If the instructor's workload allows it we would prefer not to set a maximum limitation since this could limit students who want to write an in-depth reflection essay. In general, we do not want to set a rule for failing if the word count is > 2500.
1.2 Correct template	A new template was created which offers a clear structure for the essay. It follows the four phases of reflective activity. Every phase should be explored in a dedicated chapter within the essay.	Does the essay contain all four chapters of the four phases of reflective activity?	One of the four phases is not represented in the essay or is only mentioned in two sentences.	The new template has a clear structure and allows the instructor to evaluate easily if an important part of the reflection is missing.
C2: AI Use	According to the reflection essay task, students were asked not to use AI generation tools beyond editing their text and to describe in the final passage how they used the additional tools.			As of now, no AI detection tool offers robust enough predictions and clear distinction between text generation and grammar/spelling improving AI like Grammarly. Therefore, we recommend conducting research every year before essay correction to keep up with the technical progress of AI detection.
2.1 Level of AI Use	Estimation of AI usage percentage in the essay by a detection tool.	Percentage estimation by a tool.	This alone does not lead to failure.	This criterion was separated because high AI usage does not necessarily imply inappropriate use (e.g., grammar/spell checking or vocabulary improvement).
2.2 Appropriate Use of AI	Appropriate Use of AI: checking grammar or better vocabulary Inappropriate Use of AI: text or idea generation The students must mention for which purpose they used which AI tool.	Ideally, students mention AI use honestly; alternatively, a tool or instructor judgment decides inappropriate use. We use: 1. Asking students 2. If a paragraph is high % and feels generated	Mentioned inappropriate use of AI. Alternatively: if the majority of text passages are estimated to have >90 % AI usage and appear solely generated.	Assuming not every student honestly describes their AI use; the instructor must determine whether AI use is inappropriate. This may correlate with level of AI use but does not have to, which is why these criteria are separated.
C3: Level of Complexity	Complexity refers to the complexity of reflection and thoughts expressed in the essay. Does the student describe complex learning and insights? This can be assessed with a scale where each level builds on the previous one, adding criteria.	Low: describes obvious learnings within course scope. Medium: describes somewhat complex learnings or extends course learnings with additional thoughts. High: describes learnings beyond course scope or relates insights to new contexts and extends first insight into a second.	If both C3 and C4 are considered low.	We revised "Level of Complexity" to focus on train of thought and insights rather than text readability metrics.
C4: Level of Personal Insight / Emotionality	This criterion reflects whether the student offers insights into their personal course experience and/or emotions. Measured on a scale where each level builds on the previous.	Very low: brief, matter-of-fact personal references. Low: personal examples without emotions. Medium: examples analyzed without addressing emotions. High: examples deeply analyzed with simple emotions and connections beyond course. Very high: examples with complex analysis, complex emotions, and personal connections beyond course.	If both C3 and C4 are considered low.	The scale conflates depth of analysis and expression of emotions, making levels 3 and 4 hard to distinguish.