DeThink Accessibility Toolbar
color contrast
text size
highlighting content
zoom in
< Go back to toolkit

Introduction

This technique allows for the visual organization of collected information and identification of revelations. It is important to sift through the information obtained in the empathy phase.

Saturate

The technique saturate and group is also known as Clustering, data segmentation, clustering technique, group classification technique.

Description

What is this technique/tool about

The clustering technique allows to organise large datasets, identify specific groups or segments, and analyse similar trends or behaviours, providing a better understanding and organisation of information.

There are different clustering algorithms, such as k-means, DBSCAN, hierarchical, among others, each with its own characteristics and approaches. The choice of algorithm depends on the nature of the data and the objectives of the analysis.

For which purposes is it used (why in your high school teaching)

The clustering technique is used in different contexts and disciplines due to its various applications and benefits. The following are some of the reasons why clustering is used:

  • Pattern discovery: Clustering allows the discovery of hidden patterns or inherent structures in data. By grouping similar elements into clusters, relationships and trends can be identified that are not obvious to the naked eye. This can provide valuable information for decision making and data understanding.
  • Segmentation and personalisation: Clustering is used to segment and categorise groups of individuals or entities with similar characteristics. This segmentation allows for a better understanding of the needs and preferences of different groups and facilitates the customisation of products, services, or interventions. In the educational context, student segmentation can help to design more effective teaching strategies tailored to the individual needs of students.
  • Market analysis: Clustering is used in market analysis to identify customer segments with similar characteristics and behaviour.
  • Anomaly detection: Clustering is also used to detect anomalies or unusual patterns in the data. By identifying well-defined groups, any object that does not fit neatly into any of the groups can be considered an anomaly or outlier.

In the context of secondary education, the use of clustering can help to identify different profiles of students with similar characteristics, allowing educators to adapt their pedagogical approach and provide more personalised interventions. It can also be useful for segmenting groups of students with specific needs, which facilitates the planning of appropriate support programmes and educational resources for each group.

 

Limitations

Although the clustering technique is a useful tool, it also has some limitations that are important to bear in mind:

Dependence on data quality: The quality of the input data can have a significant impact on the clustering results. If the data are incomplete, have errors or contain outliers, this can negatively affect the accuracy and consistency of the clusters obtained. It is important to perform a data cleaning and pre-processing process to reduce these effects.

Subjective interpretation of results: The interpretation of clustering results can be subjective and depends largely on the knowledge and experience of the analyst.

How to implement these technique/tools

Clustering consists of grouping items into groups with similar characteristics and is used to determine weather patterns, group items by topic or to segment customers. In short, its objective is to form closed and homogeneous groups from a set of elements that have different characteristics or properties but share certain similarities.

This technique allows for the visual organization of collected information and identification of revelations. It is important to sift through the information obtained in the empathy phase.

To develop it, we will need a wall to place post-its on.

Preparation, before the session:

  • Gather any materials or resources needed for the session, such as post-its, presentations, documents, or images.
  • Create an agenda for the session, outlining the objectives, activities, and timeline. Share this with the students in advance so they know what to expect and can come prepared.
  • Think about how you will facilitate the session and engage the students. Plan how you will introduce the activities, give instructions, and manage the time and discussions.

During the session:

  1. Information dump: also called saturation phase. It consists of transferring all the information we must post and sticking them on the wall. It is important that the phrases are constructed with subject and predicate and do not allow for ambiguity. One of the benefits of doing this is that, even if you stop working for a while, you can recover the information and understand what you wrote.
  2. Information organization: consists of organizing the post-its that have been stuck by grouping them by affinity or theme. At this point, it is important to be specific and not use categories that are too broad, such as good and bad. Keep in mind that if you are too generic, you are probably analysing the information in a shallow way.
  3. Synthesis: defining each group with a self-explanatory phrase. You must construct an explanatory sentence. The goal is to delve deeper and reflect on what is behind the information contained in each group of post-its. You will extract relevant information for the user.

Follow-up, after the session:

  • Promote discussion among students on the technique used.
  • Follow up with your students to see if they have any questions or feedback about the session. This can help you to improve future sessions and ensure that students feel supported and engaged.
  • Evaluate the session to determine what worked well and what could be improved. Use this information to adjust for future sessions.

Examples and/or testimonials

Some examples and testimonies that could illustrate the application of the clustering technique:

Example 1:

In a research project in a secondary school, students collected data on the music preferences of their classmates, such as favourite music genre, favourite artists, and frequency of listening. Using the technique of clustering, they grouped students into different segments based on their musical preferences. This allowed them to identify groups of students with similar tastes and to analyse common characteristics within each group.

Testimonial 1:

'Applying the clustering technique in our project was a fascinating experience. We were able to see how our classmates were grouped into different categories according to their musical tastes. This helped us to better understand the different social groups within our school and gave us ideas for organising musical events and activities more in line with our classmates' interests.' - Secondary school student.

Example 2:

In a social science project, students collected data on the study habits of their classmates, including variables such as time spent studying, techniques used and academic performance. Using clustering, they identified different student profiles based on their study habits and assessed the relationship between these profiles and academic success.

Testimony 2:

'The clustering project allowed us to explore how different study approaches can influence academic performance. We were able to identify groups of students with similar study strategies and observe the differences in their results. This helped us to reflect on our own study techniques and learn from the successful approaches of our peers.' - Secondary school student.

Tools needed

To develop it, we will need a wall to place post-its on.

In the context of secondary education, there are several tools that can be used to implement and apply clustering techniques.

  • Spreadsheets: Spreadsheets, such as Microsoft Excel or Google Sheets, can be used to organise and analyse data in tabular form.
  • Data visualisation tools: The use of data visualisation tools such as Tableau, Power BI or Google Data Studio can help to represent clustering results in a graphical and intuitive way.
  • Machine learning platforms: Some machine learning platforms such as TensorFlow, scikit-learn or IBM Watson Studio offer clustering functions and algorithms that can be used in the educational context.

It is important to consider the level of knowledge and technical skills of the students and the resources available in the educational institution when selecting appropriate tools. In addition, it is advisable to provide guidance and technical support to students during the use of these tools to ensure a correct implementation and understanding of clustering concepts in the context of secondary education.

Resources

Links:

< Go back to toolkit

Join Our Newsletter