PSYC100 slides 8
Welcome to PSYC100 and Retrieval Practice
Challenge Yourself! Before reviewing notes, actively try to recall information:
Recreate the timeline of the foundations of psychology.
Describe the three validities.
Check your notes for accuracy after attempting retrieval practice.
Class Tools: Get PollEv ready for later in class at www.pollev.com/punchyocean.
Today's Learning Objective
Discuss the three main research designs used in psychology.
Types of Research Methods
Psychology employs various research methods, primarily categorized into three types:
Descriptive Methods: Aim to describe what is occurring. Example: Observing and documenting how often children share toys in a classroom.
Correlational Methods: Test the relationship between different factors or variables. Example: Investigating if there's a link between hours of sleep and exam performance.
Experimental Methods: Investigate what causes a particular outcome. Example: Testing if a new medication reduces symptoms of depression.
Descriptive Methods
These methods focus on observing and documenting phenomena without manipulating variables.
Examples include:
Observational Studies: Researchers observe participants in their natural environments or a controlled setting. Example: Observing children's play behavior in a playground.
Self-reports: Participants provide information about themselves through surveys, interviews, or questionnaires. Example: Surveying students about their study habits.
Case Studies: Involve an in-depth investigation of a single individual, group, or event. Example: An in-depth study of a patient with a rare neurological condition.
Correlational Methods
Purpose: To determine if a relationship exists between two or more variables and, if so, the strength and direction of that relationship.
Operational Definition of Variables: It is crucial to define variables precisely for measurement.
Example: For variables like "Distance To Professor" and "Academic Performance," one must define how each will be measured (e.g., for "Distance To Professor" one might measure the number of rows a student sits from the front, and for "Academic Performance" it could be the final course grade expressed as a percentage on an exam).
Forms of Correlational Studies:
Form 1 Example: Investigating the relationship between how far students sit from the professor and points earned on an exam. This could reveal if students closer to the front tend to earn more points.
Form 2 Example: Investigating the relationship between IQ level and brain size. A key characteristic here is that participants are NOT assigned to groups; these are existing characteristics or groups being studied. Researchers would measure existing IQs and brain sizes and see if a pattern emerges.
Correlation Coefficient (r): A numerical value that indicates the strength and direction of a linear relationship between two variables.
The coefficient ranges from -1.0 to +1.0.
-1.0: Perfect negative correlation. Example: As hours of exercise increase, body fat percentage perfectly decreases.
-.7: Medium negative correlation. Example: As stress levels increase, overall well-being moderately decreases.
0: No correlation. Example: There is no consistent relationship between shoe size and intelligence.
+.5: Medium positive correlation. Example: As study time increases, exam scores moderately increase.
+1.0: Perfect positive correlation. Example: As the number of hours worked increases, total wages earned perfectly increase.
Scatterplots: Visual representations of the relationship between two variables.
Scenario 1: If a researcher finds that the more students engage in metacognition, the higher their grades are, this would be best depicted by a medium to perfect positive correlation scatterplot. The dots on the scatterplot would generally trend upwards and to the right.
Scenario 2: If a researcher finds that the less often parents used diapers, the less times their babies had urinary tract infections, this indicates a positive correlation. As diaper use frequency decreases, UTI frequency decreases, suggesting a direct relationship between the two factors (or an inverse relationship between frequency of diaper use and frequency of UTIs - a positive correlation where both variables move in the same direction). If you plot 'diaper changes per day' on one axis and 'UTI incidence per month' on the other, a decrease in one would correspond to a decrease in the other, showing a positive trend on the scatterplot.
Why Correlation Does Not Equal Causation: This is a critical principle in research design.
Directionality Problem: Correlation alone cannot determine which variable causes the other.
Example: Does sitting closer to the professor cause higher exam points, or do students who perform well on exams choose to sit closer? It's impossible to tell from correlation alone.
Third Variable Problem: An unmeasured, confounding variable might be responsible for the observed correlation between two variables.
Example: The correlation between distance from the professor and exam points might be due to a third variable, such as a student's motivation or study habits, which influences both where they sit and their performance.
Another Example: There is a correlation between the homicide rate and ice cream sales. However, neither causes the other. The third variable is heat (temperature). Higher temperatures lead to both increased ice cream sales and, unfortunately, often an increase in certain types of crime.
Experimental Methods
Purpose: To investigate a cause-and-effect relationship between variables.
Key Characteristics:
Experimenter Manipulation: The researcher actively manipulates one or more variables. Example: A researcher might decide to give one group a new drug and another group a placebo.
Random Assignment: Participants are randomly assigned to different groups (experimental or control/comparison) to ensure that groups are equivalent at the start of the study. This is crucial for inferring causality. Example: Flipping a coin to decide if a participant receives the experimental treatment or the control treatment.
Components of an Experiment:
Independent Variable (IV): The variable that the experimenter manipulates or changes. Each group is exposed to a specific condition of the IV. Example: In a study on study techniques, the IV would be the type of study technique used.
Dependent Variable (DV): The variable that is measured. It is the outcome believed to be affected by the IV. Example: In the study on study techniques, the DV would be the scores on a subsequent exam.
Experimental Group: The group of participants exposed to the condition of the IV being tested (the treatment). Example: The group of students using the new, experimental study technique.
Control or Comparison Group: The group of participants exposed to a different condition of the IV, often a baseline, placebo, or absence of the treatment. This group serves as a comparison for the experimental group. Example: The group of students using traditional study methods or no specific technique.
Comprehensive Example: To test if a new study technique (IV) improves test scores (DV), researchers might randomly assign one group of students (experimental group) to use the new technique, while another group (control group) uses traditional study methods. Both groups then take the same test, and their scores (DV) are compared.
Process: A population of interest is sampled, and then the sample participants are randomly assigned to either the experimental group or the control group within the study (refer to Figure 2.15 in the textbook).
Important Reference: Ensure to review the text related to Figures 2.27 and 2.28 for further understanding of experimental design.
Exercise: Analyzing Experimental Design
Watch a clip from a Brain Games episode (www.pollev.com/punchyocean).
Note: While it demonstrates a study, it is a television show and may not be as controlled as a real scientific study.
Answer the following questions based on the clip:
What is the Independent Variable (IV)? How was it manipulated?
What is the Dependent Variable (DV)? How was it measured?
What are some potential confounds (uncontrolled variables that could affect the outcome)?
What kind of validity is affected by having these confounds?