Observational Experiments
1 What are they?
An observational experiment is an experiment that students perform when they are investigating a new phenomenon. Thus, they do not make predictions or have expectations about its outcome. Students need to collect data, analyze them and find a pattern in the data.
They then need to explain the reasons for the pattern (if applicable), and/or construct a qualitative or quantitative relationship.
2 Why do you want to use them?
Observational experiments have the following desirable features:
• Concrete experiences
These experiments provide concrete experiences as part of the learning cycle.
• Open-mindedness
Students learn to learn to approach data without knowing whether they are right or wrong.
• Decision-making
Students learn to make decisions about representing data, considering outliers, deciding what pattern they want to look for.
• Epistemic cognition
Students learn to deal with situations when there is no right answer. They learn how to choose a productive way to investigate complex phenomena.
• Real data
Students learn to deal with the complexities of real data.
3 How and where do you use them?

Observational experiments can be used in the following contexts:
• In a lecture while developing ideas in a new topic.
The instructor performs the experiment, students record data, decide what variables are important, and try to find patterns in the data by plotting graphs.
• In the lecture, laboratory or recitation/workshop.
The data from such experiments (performed by someone else) are provided. Students analyze them and look for patterns.
• In the lecture, laboratory or recitation/workshop.
The data and analysis are provided. Students construct explanations for the trends in the data.
• In the laboratory. Students perform these experiments in the lab (before learning about them in lecture), collect and analyze data, find patterns, and construct explanations or mathematical relationships to describe the patterns.
• As a homework problem. Data for an experiment are provided and students are asked to analyze the data and find patterns in them.
4 What are some types
Observational experiments can be used to introduce any new area of physics. They can be either qualitative or quantitative. As described in the previous section, some or all parts of an observational experiment can be used for example, students can collect and analyze data, and/or find patterns, and/or construct explanations.
5 How do you score them?
We present an example of an observational experiment, a model solution and an example of student work. After this we present the rubrics we use, the scores on various abilities, and reasons as to why particular scores were given.