Lesson 7: Observational Studies vs. Experiments

Objective:

Students will learn how observational studies differ from experiments, and will classify different research scenarios based on which method would be most appropriate. They will also learn about the roles of ethics, cost limitations, and feasibility when deciding between the two data collection methods.

Materials:

What Should We Do? handout (LMR_3.3_ObsStudies vs Experiments)

Vocabulary:

ethics cost limitations feasibility

Essential Concepts:

Experiments are not always possible because of various factors such as ethics, cost limitations, and feasibility.

Lesson:

Remind students that in observational studies, we can never randomly assign subjects to treatment and control groups. Conversely, in experiments, we always need to have random assignment into these groups.
Pose the following question to students: Why can’t we just always do experiments? Have students discuss this question in their student teams and write down a few responses in their DS journals.
Inform students that a researcher wants to perform studies to answer the research questions below. In teams, have students come up with reasons for why an experiment would not be possible for each scenario.
1. Does smoking cause lung cancer? Unethical. You cannot make people smoke cigarettes and then see if they have lung cancer later in life.
2. Does drinking water from Mars keep you healthier than drinking water from Earth? Cost. It would be incredibly expensive to design a space shuttle that can successfully transport people to Mars and have them live there for an extended period of time and most researchers would not have the funding to do this.
3. Do people with higher IQ scores score better on the SAT than people with lower IQ scores? Not feasible/not possible. You cannot randomly assign IQ scores to people because it is a measurement based on aptitude.
Select three teams and assign a scenario above to each team. Ask each team to report out on their assigned scenario. As teams share, be sure to discuss the following issues regarding why we cannot always to experiments:
1. Ethics: Sometimes, experiments cannot be performed because it would be unethical to give certain treatments to subjects. For example, we could not inject an HIV infection into participants because the long-term effects might lead to death.
2. Cost Limitations: Sometimes, experiments would be very costly and much too expensive to perform. Some possibilities could be with technology.
3. Feasibility, impossible to randomize: In certain cases, you cannot perform an experiment because it is impossible to randomly assign people to particular groups. For example, you cannot assign a gender to a person.
Distribute What Should We Do? handout (LMR_3.3). In teams, students will identify whether the research question could best be answered via an experiment or an observational study.
Once all student teams have completed the handout, assign one research question to each team to report out. As each response is shared, conduct a whole-class discussion to compare which data collection method was most appropriate for each research question. Ensure everyone understands the reasons each method was chosen before moving on to the next scenario.

Note: Page 2 of the handout is an answer key for teacher reference only!

LMR_3.3
Next, student teams will generate three research questions on their own. They need to identify the best data collection method for answering their question and should provide an explanation. At least one of the three research questions should use an observational study for data collection.
Using a share-out strategy, have the reporter of each team share one of their investigation questions with the rest of the class.

Class Scribes:

One team of students will give a brief talk to discuss what they think the 3 most important topics of the day were.