Skip to content

Lesson 2: What Is an Experiment?

Lesson 2: What is an Experiment?

Objective

You will learn about the elements of an experiment and the meaning of "causation". You will learn to distinguish claims of causation from claims of association.

Materials

  1. MythBusters videos, Is Yawning Contagious?

    1. MythBusters video, Is Yawning Contagious?, Part 1

    2. MythBusters video, Is Yawning Contagious?, Part 2

    Note: If video is not found using link, please use a search engine (e.g., Google Video) and type “MythBusters Is Yawning Contagious” to find it. The clip is a little over 5 minutes in length.

Vocabulary

causation, association, correlation, experiment, research question, subjects, treatment, treatment group, control group, random assignment, outcome, statistic

Essential Concepts

Essential Concepts

Science is often concerned with the question "What causes things to happen?" To answer this, controlled experiments are required. Controlled experiments have several key features: (1) there is a treatment variable and a response variable, and we wish to see if the treatment causes a change that we can measure with the response variable; (2) There is a comparison/control group; (3) Subjects are assigned randomly to treatment or control (randomized assignment); (4) Subjects are not aware of which group they are in (a 'blind'). This may require the use of a placebo for those in the control group; and (5) those who measure the response variable do not know which group the subjects were in (if both 4 and 5 are satisfied, this is a "double blind" experiment).

Lesson

  1. What is the difference between causation and association? Causation means there is a cause and effect relationship between variables. Association, or correlation, means that high values of one variable tend to be associated with high values of the other (or high values tend to be with low values). Based on these descriptions, let's determine which claim below is attributed to causation and which one is attributed to association (also referred to as correlation):

    1. Heat makes water boil. (Causation)

    2. Blanket sales in Canada rise as brush fires in Australia rise. (Association)

  2. In statement (a) the relationship between heat and water is that heat causes water to boil. While it may be true that as blanket sales rise, the occurrence of brush fires also rises, it doesn't mean blanket sales cause brush fires. So in statement (b) there isn't necessarily cause-and-effect in play. Blanket sales rise in the cold winter in Canada, which coincides with summer in Australia (when brush fires tend to occur there).

  3. Read the following headlines:

    1. Stop Global Warming: Become a Pirate

    2. Lack of sleep may shrink your brain

    3. Early language skills reduce preschool tantrums

    4. Dogs walked by men are more aggressive

  4. Analyze each headline (3a-3d) by responding to each of the following questions in the boxes, then click to check your work:

    What is the headline implying with its wording? 3a is implying that you can stop global warming by becoming a pirate, 3b is implying that it’s possible to shrink your brain if you aren’t getting enough sleep, 3c is implying that having early language skills will decrease preschool tantrums, and 3d is implying that dogs are more aggressive when they’ve been walked by men.

    Are the headlines implying causation or association? Causation means there is a cause and effect relationship between variables, whereas association or correlation means that high values of one variable tend to be associated with high values of the other (or high values tend to be with low values). 3a, 3c and 3d are implying causation, and 3b is implying association.

    How can you tell the difference between causation and correlation? What words stand out in these headlines? Some possible terms for causation include: cause, increase/ decrease, benefits, impacts, effect/affect.

    Some terms for correlation include: get, have, linked, more/less, tied, connected.

    In 3a, “become” stands out; in 3b, “may” stands out; in 3c, “reduce” stands out; in 3d, “are” stands out.

  5. Consider the headline: "Is Yawning Contagious?” Respond to each of the following questions in the boxes, then click to check your work:

    What is the headline implying with its wording? That yawning may cause other people to yawn.

    Is it implying causation or correlation? How do you know? Causation, because “contagious” yawns means that you are yawning because someone else has yawned.

    How can we determine if this is true? By running an experiment.

  6. The show Mythbusters investigated whether or not yawning is contagious. Click on the links below to view a video clip of their investigation.

    1. Part 1

    2. Part 2

    As you watch, focus on the guiding questions below and jot down notes as needed in your IDS Journal:

    1. How did the MythBusters design the investigation?

    2. What steps did they take?

    3. How is this different than your answer to the third question in #5?

  7. The Mythbusters have just conducted an experiment, which is a data collection method. In this lesson and in the ones that follow, you will understand what conducting an experiment requires.

  8. Below is a list of elements that are found in experiments. In parenthesis, you will see how each element was addressed in the Mythbusters video clip.

    1. Research question: The question to be answered by the experiment (Is yawning contagious?)

    2. Subjects: People or objects that are participating in the experiment (The 50 adults)

    3. Treatment: The procedure that is assigned to a group of subjects (Kari yawned before the subject entered the room)

    4. Treatment group: The group of subjects that receive the treatment (Two out of every three subjects who were placed into rooms – yawn from Kari)

    5. Control group: The group that does not receive a treatment (One out of every three subjects who were placed into rooms – no yawn from Kari)

    6. Random assignment: Subjects are randomly assigned to either the treatment or control group (Two out of every three subjects received the treatment)

    7. Outcome: The variable that the treatment is meant to influence. (Whether or not a person yawned)

    8. Statistic: A method for comparing the outcomes of the control and treatment groups is needed. You need to have a statistic as a method for comparing the outcomes of the control and treatment groups. In this case, the MythBusters used the difference between the percent of subjects that yawned in the treatment group was 4% higher than the control group.

    Note: In this experiment, random assignment was not used (or if it was, we were not told). Also this experiment, and in those found in the IDS curriculum, we use a treatment and a control group. However, a control group is not a necessary element of an experiment. Sometimes it is more appropriate to have two treatment groups with no control group (e.g., medical professionals testing different doses of drugs). The effect that is being studied will dictate whether to feature a control group or not.

  9. Write down your thoughts about the questions in the boxes below in your IDS Journal, then click to check your work:

    Why did the MythBusters follow all of these steps to design their experiment? In order to determine if watching someone yawn can cause you to yawn.

    We don't know how MythBusters chose who would be in the treatment group and who would be in the control group. Suppose that the people who showed up first, early in the morning, were assigned to the treatment group, and the last few people, later in the day, ended up in the control group. Would you believe in the conclusions? No, because the two groups were different. The first group might have been sleepier, and therefore more likely to yawn anyways. This, another explanation for the cause-and-effect, is a confounding variable.

    In order to make the two groups as similar as possible, experimenters usually assign subjects at random. How might we randomly assign about half of the subjects to the treatment and half to the control? We might flip a coin, and those who get Heads go to Treatment.

    Why would random assignment improve the MythBusters study? Because then the two groups would be more similar, so we wouldn't have a confounding variable to worry about.

  10. Important point: Without random assignment, we cannot determine causation because we are not comparing two similar groups.

Reflection

What are the essential learnings you are taking away from this lesson?