True Experimental Design

Scientists in virtually every field of study utilize experiments to objectively answer research questions.
An experiment is the primary method used to investigate causal relationships, such as whether caffeine has a specific effect on student behaviors and performance within a classroom setting.
This lesson aims to cover all critical aspects of experimental design, including variables, group categorization, assignment methods, and sampling techniques.

The first step for any researcher is to identify the specific variables under study.
Variables are typically framed within a hypothesis. For example, a researcher might hypothesize that when students consume caffeine, their performance on tests is affected.
- One hypothesis could be that caffeine increases test performance by making students less sleepy and more focused.
- An alternative hypothesis could be that caffeine decreases test performance by making students jumpy and hyper.
There are $2$ primary types of variables in an experiment:
The Independent Variable (IV):
- Definition: This is the variable controlled or manipulated by the experimenter. It is the factor that creates the different groups in a study.
- Role: It acts as the "cause" in the cause-and-effect relationship.
- Temporal Order: The independent variable always occurs first.
- Example: In a caffeine study, the independent variable is the caffeine itself (e.g., giving one group caffeine and the other water).
The Dependent Variable (DV):
- Definition: This is the outcome variable that is measured by the researcher.
- Role: It acts as the "effect" in the cause-and-effect relationship. It is the variable that the researcher expects to change based on the manipulation of the independent variable.
- Temporal Order: The dependent variable always occurs second.
- Example: In the caffeine study, the dependent variable is the test performance.

An experiment requires at least $2$ groups to allow for comparison. If a researcher only used $1$ group (e.g., giving caffeine to every student and measuring how many times they could jump a rope), there would be no way to determine if the performance was higher or lower than it would have been without the caffeine.
The Control Group:
- Definition: This is the natural or baseline group that provides a standard for comparison.
- Environment: This group does not receive any change to their natural state.
- Example: Children who drink only water in the caffeine experiment.
The Experimental Group:
- Definition: This is the group that receives a specific change or manipulation to their natural environment.
- Example: Children who drink caffeine in the caffeine experiment.
The Placebo Effect:
- This occurs when participants change their behavior simply because they expect a change to happen, rather than because of the actual manipulation.
- To avoid this, researchers may use a "blinded" approach. For example, giving all children soda where half ( $\frac{1}{2}$ ) have caffeine and half ( $\frac{1}{2}$ ) are caffeine-free, without telling them which is which. A good control group helps mitigate this problem.

Random Assignment:
- Definition: A process where every participant in the study has an equal chance of being placed into either the control group or the experimental group.
- Method: Decisions can be made through random methods, such as flipping a coin.
Purpose of Random Assignment:
- It ensures that the groups are as identical as possible in every way except for the independent variable.
Consequences of Poor Assignment (Confounding Variables):
- If a researcher assigned all boys to the caffeine group and all girls to the no-caffeine group, and then found that the caffeine group was better at jumping rope, they could not conclude caffeine was the cause. It might simply be that boys are better at jumping rope.
- Without random assignment, other factors (extraneous variables) could explain the difference between groups. Only with random assignment can a true cause-and-effect relationship be established with certainty.

Population:
- Definition: The entire, relatively large group of people that the researcher is interested in studying.
- Range: If studying the effects of caffeine on children, the population could be every child in the world, or more specifically, every child in a particular school.
Sample:
- Definition: The specific, smaller group of people who are actually participating in the study.
- The sample is always smaller than the population.

Convenience Sample:
- Definition: A sample chosen from people who are readily available and willing to participate (e.g., friends, family, or volunteers from a personal Facebook page).
- Disadvantage: It is often not representative because it only captures a certain "type" of person.
Clustered Sample:
- Definition: A specific type of convenience sample based on a particular geographic area.
- Example: Only sampling people in California. This may not represent people in Iowa, who might behave differently.
Random Sample:
- Definition: A sample where every person in the larger population has an equal chance of being selected for the sample.
- Advantage: It is highly likely to be representative, meaning important variables (e.g., age, grade level, geography, ethnicity) present in the population are equally present in the sample.
Stratified Random Sample:
- Definition: A structured yet random approach where researchers identify important variables in advance (e.g., gender), divide the population based on those variables, and then randomly choose an equal number of people from each group.
- Example: Randomly choosing an equal number of boys and girls to ensure the sample is balanced for gender comparison.

A "true experiment" requires the careful management of the following:
- Independent Variable: Controlled cause.
- Dependent Variable: Measured effect.
- Control Group: The baseline/natural group.
- Experimental Group: The group receiving the manipulation.
- Random Assignment: Required to ensure group identity and establish causation.
- Representative Sampling: Ideally achieved through a Random Sample to ensure findings can generalize to the larger population.
Adhering to these principles allows researchers in psychology and other fields to create rigorous, valid studies.