Experiments

Modified: 2023-08-15 (8:35 PM CDST)

• Experiments provide the only method by which allow the researcher can specify cause and effect.
  • Experiments are set up with minimum of two groups.
    • One group, called the control group, is not given the experimental treatment by the experimenter.
    • The other group, called the experimental group, is given some procedure, treatment, or substance by the experimenter.
  • Everything else about the two groups is as similar as possible.
  • More than one procedure, treatment, or substance can be tested at the same time by adding additional experimental groups.
• Let's look at two examples:
  • The first one looks at toothpaste use and number of cavities. The second looks at the effects of a sleeping pill.
  • If nothing else they each illustrate several important concepts:
    • independent variable(s) or IV(s)
    • dependent variables(s) or DV(s)
    • operational definitions
• Dependent variables (DVs) are easier to find in an experiment
• They are the measurements that ALL respondents are subjected to.
• In the first example that follows the dependent variable will be number of cavities
• In the second experiment the three dependent variables will be:
  • hours of sleep
  • quality of sleep
  • latency of sleep
• All will be explained below. For now, remember that the value of the dependent variable will change as a consequence of the independent variable.
  • Independent variables (IVs) are a little harder to understand or to find in an experiment, especially early in your psychological careers.
    • Here are several ways to think of an IV:
      • It is the reason for doing the experiment. In the first example using or not using toothpaste, or "dental care," is the independent variable. The research question is: "Does using toothpaste cause fewer cavities?" A properly run experiment should be able to answer that cause and effect question.
      • In the second example "dosage" of the sleeping pill is the independent variable. The two dosages are one sleeping pill or a pill that looks like the sleeping pill but does not have any active ingredients in it. (Yes, it's a placebo)
      • Another definition is that the IV is the only uncontrolled variable in the experiment.
      • Experiments attempt to keep all other variables constant while allowing the levels of the IV (toothpaste or no toothpaste) be uncontrolled.
• Operational definitions are common in science.
  • Operational definitions provide a set of procedures that define a concept.
    • For example, if I ask if you are hungry you will likely give me an answer based on your feelings.
    • And, if I ask 100 people they will do the same, but I'll have no way of comparing all of those internal feelings of hunger.
    • So, what should I do?
      • One way is to change the question: "When was the last time you ate? is such a question.
      • This question operationalizes hunger into a measurable variable where having eaten two hours ago might equal "not very hungry" and where having eaten 24 hours ago might equal "very hungry."
    • Notice that the operational definition for hunger here is: hours after eating.
      • WARNING: other research shows that hunger feelings do rise day after day, but later they plateau, later still they decline. So, this definition for hunger only works over the span of a few days. Those feelings of hunger will actually disappear.
• For the first experiment all participants must have "healthy teeth" which is operationally defined as: having their own biological teeth, (no false teeth, no implants), having healthy teeth (not rotten ones), and zero cavities (which a dentist will assure prior to the experiment.)
• For the second experiment, three operational definitions will be required.
  • For hours of sleep, time will be measured by EEG sleep responses (see chapter 3)
  • for quality of sleep via a questionnaire after they wake
  • for latency of sleep (how quickly they fall asleep) by measuring the time on the EEG between an active waking EEG and a sleep EEG.
• The first experiment is simpler:
  • We recruit a large number of people willing to participate.
  • Divide them at random into two groups (more on randomization later)
  • Eliminate anyone whose teeth fail to meet the operational definition abovef
  • Fill all cavities so that at the beginning of the experiment everyone has zero cavities.
• Next:
  • eveyone is given the same toothbrush
  • taught how to brush the same way
  • agree to brush three times a day
  • One group does all that with toothpaste
  • The other only with water.
  • (Note that the IV = Dental Care with two levels: brushing with toothpaste and brushing with water.)
  • So, the only thing different, the only thing left to its own, or uncontrolled are the two levels of the IV.
• The hypothesis here is:
  • Brushing with toothpaste will lead to fewer cavities.
  • If the toothpaste using group ends up with fewer cavities it will be ONLY due to the difference in the two levels of the IV because all other variables were the same in both groups.
• Now, let's move to a more complicated example.
  • Suppose we have created a new drug, a sleeping pill. We call it Nox-Out©.
  • Now, we are interested in marketing it by saying that it is better at putting people to sleep than other sleeping pills. So, we decide to conduct an experiment.
• First, we need some terminology.
  • We can name as the control group the group that does not get the procedure, treatment, or substance (a level of the IV).
  • In the Nox-Out© case the dose of Nox-Out© is the substance, here, one level of the independent variable.
  • So, now we can say that the control group does not get the Nox-Out©.
  • For contrast, the experimental group does get the Nox-Out©.
  • Let's call the IV "Sleep substance" with two levels = one dose of Nox-Out© vs No dose of Nox-Out©. (Later we'll add the placebo as another control)
• Finally, the dependent variable is how we measure the effects of the independent variable.
• Both groups will be measured in terms of the dependent variable.
• In this experiment, one good dependent variable might be hours of sleep after taking Nox-Out©.
• We could have additional DVs: quality of sleep and latency to sleep, as noted above. (And other hypotheses as well)
  • There will be three hypotheses, one for each DV:
    • Nox-Out© makes people sleep longer.
    • Nox-Out© makes people sleep betterl
    • Nox-Out© makes people fall asleep faster.
• Can you see that all three hypotheses are important?
  • No one will buy a sleeping pill that makes them sleep better if their quality of sleep goes down.
  • All three hypotheses should be supported.
• Let's make experimental mistakes just to see what might happen.
  • The first bad experiment.
    • Suppose the subjects in the control group are all between 60 and 80 years of age, and the subjects in the experimental group are all between 20 and 30 years of age.
    • When we measure both groups using the dependent variable of hours of sleep, we may find that the young group sleeps longer.
    • But, did that difference in the dependent variable come from the independent variable or from the age difference?
    • Here, we cannot tell.
    • When two or more factors contribute to a dependent variable difference, it is called a confound.
    • Confounds are always a problem in experimentation, and they are the main reason that the study of experimental design is important.
    • Good experimental designs eliminate confounds.
  • To fix this bad experiment both groups must be the same regarding their age
  • Here is another bad experiment.
    • Now let's house the control group in the new Hilton Hotel
      • it's quiet
    • Let's house the experimental group in the El Sleazo Motel
      • the one with the neon light that flashes all night.
      • the one by the busy highway with trucks going by all night.
    • Now our dependent variable difference may show that Nox-Out© does not work very well.
    • Again, the problem is one of experimental design.
    • Subjects just find it more difficult to sleep at the El Sleazo, even when they have taken Nox-Out©.
    • This example, too, is a confound.
    • To fix this bad experiment both groups should sleep in similar environments
  • A final bad experiment could be one
    • In which we give Nox-Out© to the experimental group at 10:00 a.m. and then put them to bed; we put the control group to bed at 11:00 p.m.
    • Again, the dependent-variable differences will lead us to believe that Nox-Out© does not work very effectively.
    • To fix this experiment both groups should be awake the same amount of time.
• The message here is pretty obvious.
  • Good experiments should differetiate the experimental and control groups only by giving or not giving the independent variable.
  • So, to properly test Nox-Out©, both groups should be of the same age, sex, and degree of sleepiness.
  • Also, they should be tested in the same environment (e.g.,. the Hilton, preferably).
• When all of these details are taken care of, and when a difference in the dependent variable exists, then the experimenter can say it was the independent variable that caused the difference.
  • The ability to make such statements is the main benefit of the experimental method.
  • However, just because a study is an experiment does not give it that power.
  • There have been plenty of bad experiments like those described above.
  • So, a sophisticated student will look beyond the dependent variable to the design and the conduct of the experiment in order to determine the validity of the conclusion drawn from the data.