Data Detectives at Work

Data exploration of multivariate data on the media use of young people

Core idea

This teaching unit is about data science content for grades 8-10 (age 14+). Using a fictitious example of an online platform that wants to tailor advertising to young people, pupils are motivated to search for traces and patterns in a data set as data detectives in order to advise the online platform.

Worksheets, PowerPoint presentations, instructions, the YOU-PB dataset, a list of variables, and an overview of the lessons are available. The CODAP software (codap.concord.org), which is freely available on the Internet, is used for data analysis.

The project works with data from over 1000 pupils who have provided information on many varia-bles in the leisure and media sector (YOU-PB stands for "youth media usage in the area of Pader-born”). The data set for this series of lessons is available in different versions. On the one hand, a reduced version with 50 variables is available. On the other hand, the full version with over 160 vari-ables can be used in lessons. It is recommended to use the data set with the reduced number of vari-ables for better clarity. Multivariate and interesting discoveries are possible here! However, the teacher has the option of differentiating according to the ability of the class/course, or to provide an internally differentiated offer for particularly able pupils (see below).

The link to the data set used in the project (50 variables):

https://tinyurl.com/you-pb-50en

The series of lessons comprises approx. 8 lessons. In the first lessons, the students are introduced to the data and working with CODAP. The core of the series of lessons is independent data science pro-ject work by the students in lessons 5 and 6 with presentations in lesson 7. The project work takes place in small groups in which the students work independently as data science experts with the data set. For this purpose, they assign themselves to four groups of the YOU-PB data, analogous to the customer wishes of the online platform (this assignment takes place in lesson 4):

  • Customer 1 wants to promote TikTok,
  • Customer 2 wants to promote LetsPlay_YouTube videos,
  • Customer 3 would like to advertise online newspapers,
  • Customer 4 would like to advertise fixed game consoles.

In the last hour, there is a reflection on the procedure for data exploration and personal and social aspects can be discussed. In addition, the topic of data cleansing, which takes up a lot of time in the work of real data scientists, can be addressed here as an excursus and worked on in CODAP.

Target group

Mathematics from grade 8 (all school types)

Computer science from grade 8 (all school types)

Recommendation: From grade 9 on

Prior knowledge

Percentages

Time scope

 8 to 10 lessons of 45 minutes each

Goals

Statistical learning objectives
  • Learners explore multivariate data
  • Learners carry out group comparisons
  • Learners use row, column, and cell percentages and interpret the results
Computer science learning objectives
  • Learners gain an understanding of the structure and functioning of IT systems
  • Learners interpret data as information
  • Learners discuss social responsibility and the impact of the use of data
Media learning goals
  • Learners use digital tools to analyze data
  • Learners create presentations and practice presenting

Lesson overview

Lesson 1: Introduction to the project and the data

CODAP: Common Online Data Analysis Platform

COCODAP (codap.concord.org) is a free, browser-based software designed for use from grade 3 onwards. It supports students in learning and using statistics and data science. No registration is required to use the platform.

. Toolkit dieser Website gibt es viel Lernmaterial zur Arbeit mit CODAP.

Susanne Podworny's YouTube channel offers further tutorial videos on how to use CODAP (in German): https://youtube.com/playlist?list=PLzhRG7IPqbqvKH5X6TIIfF8IzATgxHjQP&feature=shared

CODAP is available in many languages. If the language is not set to your preferred language, the language can be changed at the top right.

The YOU-PB data

The YOU-PB data used here comes from a Paderborn replication study of the official JIM study, which replicates these results locally and makes them available for analysis. Since 1998, the JIM study has been conducting annual surveys on the media use of 12- to 19-year-olds in order to identify trends and developments in the digital behaviour of young people and to derive new impulses for education and culture.

The data is available in CODAP in two English versions (and four German versions).

Wir empfehlen Variante 1 für den Unterricht.

Versionens of the data:

[German only] 50 variables, binary coded https://tinyurl.com/you-pb-50binaer
(Long: https://codap.concord.org/app/static/dg/en/cert/#shared=httpscfm-shared.concord.org6W1Sp7Zn06C6ad9ED53mfile.json) https://codap.concord.org/app/static/dg/de/cert/index.html#shared=https%3A%2F%2Fcfm-shared.concord.org%2F9IuhADJvA994rSs6MLct%2Ffile.json)

English: 50 variables, up to 7 categories https://tinyurl.com/you-pb-50en
(Long: https://codap.concord.org/app/static/dg/en/cert/#shared=httpscfm-shared.concord.org6W1Sp7Zn06C6ad9ED53mfile.json) https://codap.concord.org/app/static/dg/de/cert/index.html#shared=https%3A%2F%2Fcfm-shared.concord.org%2FH94YlIcxyFEYY9Ou8ZeL%2Ffile.json)

English: 160 variables, up to 7 categories: https://tinyurl.com/you-pb-en160
(Long: https://codap.concord.org/app/static/dg/en/cert/#shared=httpscfm-shared.concord.org6W1Sp7Zn06C6ad9ED53mfile.json) https://codap.concord.org/app/static/dg/de/cert/index.html#shared=https%3A%2F%2Fcfm-shared.concord.org%2FDohqN1dGy7r2mf9iqL2j%2Ffile.json)

[German only]: 160 variables: binary coded: https://tinyurl.com/you-pb-160binaer
(Long: https://codap.concord.org/app/static/dg/en/cert/#shared=httpscfm-shared.concord.org6W1Sp7Zn06C6ad9ED53mfile.json) https://codap.concord.org/app/static/dg/de/cert/index.html#shared=https%3A%2F%2Fcfm-shared.concord.org%2FstEcYRejbsfAPmTy9a26%2Ffile.json)

Further information

Opportunity for differentiation

The series of lessons is designed to work with a "small" data set. This contains 50 variables and offers plenty of opportunities for exploration.

For particularly motivated students, it is also possible to work with the large data set instead, which contains all 160 variables that were collected in the sur-vey. However, this requires a high level of commitment and good work with the list of variables on the part of the students!

Another possibility for differentiation is to have particularly motivated students work with the "normal" variables with all seven values. The standard case should be working with binary variables, i.e. variables that have previously been recoded by the pupils, as described in Instruction_recoding in lesson 2+3.

How to deal with the tasks

In the series of lessons, a lot of work is done with worksheets. In order to document the learning process and at the same time keep motivation high, the tasks can also be worked on directly in a PowerPoint presentation. New tasks can be worked on new slides and at the end, results can be taken from the various lessons to create the final presentation.

Participation in the survey (German only)

Anyone who would like to take part in the survey with their class is welcome to do so. The data will be collected completely anonymously (clarify with the school management if necessary). The data will then be included in a new edition of the data set every year, which will remain accessible via a CODAP link.

Students can take part in the survey via this link (note: 161 questions!, didicate sufficient time): http://go.upb.de/JIM-Umfrage http://go.upb.de/YOU-PB-Umfrage

Kurzübersicht über die Inhalte der Unterrichtsstunden

Citation:

Podworny, S., Frischemeier, D., Stroop, D., Biehler, R., Fleischer, Y., Schulte, C., Höper, L. & Hüsing, S. (2025). Datendetektiv:innen bei der Arbeit. . https://www.prodabi.de/materialien/datendetektiv_innen/

Published on 13.02.2025

Version: 4

License note:

Creative Commons Attribution-ShareAlike (CC BY-SA 4.0)

Authors: Susanne Podworny, Daniel Frischemeier, Dietlinde Stroop,
Nach oben scrollen