MHS AP Psych Modules 4-6

Why are science-based answers more valid than those answers based on common sense?

Common sense answers are based on casual observation rather than scientific experiments that have been peer reviewed and replicated. Common sense answers also may not actually be right.

Define and provide an example of the following cognitive biases: Hindsight Bias, Overconfidence, Perceiving order in random events, Confirmation Bias.

Hindsight bias: People tend to believe that they were right all along, even if they weren’t that sure. Ex. Saying “I told you so” after a sports team won, even if you weren’t sure

Overconfidence: People tend too think they are more skillful at something than they actually are. Ex. Thinking you can beat a pro athlete in their sport.

Perceiving order in random events: Over a large enough sample, order tends to emerge from the chaos. Ex. Flipping a coin 100 times will naturally result in a long streak of heads

Confirmation bias: Seeking out only evidence and experiments that support your viewpoint rather than deny it. Ex. Using evidence that supports veganism even if their is evidence suggesting otherwise.

How are theories and hypotheses different?

A hypothesis is an educated guess as to what researchers/people expect to find. Must be falsifiable

A theory explains behavior or events by offering ideas that organizes information. Comes from research.

Theories often come about because of hypotheses.

Falsifiable hypothesis or not.
1. If a person dreams about the future, it is a sign that they have psychic powers.

2. Wearing blue shirts improves test performance

3. Love always finds a way

1. Not falsifiable 2. Falsifiable 3. Not falsifiable

2. Explain how the following concepts are used in psychological research: Operational definitions, Replication, Peer review, Appropriate Representation.

Operational definitions: Defining variables you can observe/measure/and are specific. Used for replication.

Replication: Usually if a study finds a new conclusion, it needs to be replicated for his credibility to increase.

Peer review: Study is checked by experts to review for flaws

Appropriate representation: No sampling bias is going on, and the sample should accurate reflect the population.

How do theories advance psychological sciences?

They may be refined/changed as a result of additional experimentation. May be replicated which enhances their credibility

What are three ways to test hypotheses?

Through observation: Case studies, naturalistic observation, and surveys

Correlational Research: Associating different variables together

Experimental research: Manipulating variables to see what the effect is.

1. When doing research, methodology is divided into two categories: experimental and nonexperimental research. In the table below, describe these different types of research as well as examine the advantages and disadvantages of each.

Case Study: Looking at a rare event/condition in depth and seeing what conclusions can be drawn and how generalizable they are. Looking at a rare disease to see if information can be gained. Hard to replicate, and only relevant to a specific group

Naturalistic Observation: Observing people/animals in their natural state. Non invasive, provides info for future research, but can’t explain research

Meta analysis: Analyzing the results of many studies together. More comprehensive view, but can be compicated and time consuming.

Qualitative or Quant?

1. Qual

2. Quant

3. Quant

4. Qual

5. Qual

6. Quant

Describe the use of a survey to collect data?

Survey looks less deeply and requires people to answer questions. Wording is important

Survey data problem examples:

Framing: Don't you agree that Mansfield High School should adopt a no homework policy?

Self report bias: How many surveys have you completed this year

Social desirability issue: Have you ever cheated on a test

1. People might get confused on how the question is framed, because it is “don’t you agree” they might say yes

2. Memory isn’t 100% accurate

3. People want to appear good and this hurts their image

When doing psychological research what is meant by population? What is a sampling bias and how do you avoid it (representative sample and random sampling.

Population: All those in a group being studied. Study in sleep for teenagers would be teenagers

Sampling bias: A flaw in how people are sampled. A large unrepresentative sample isn’t better than a small one

Representative sample: A sample that accurately represents the population.

Random sampling: Every person has the same chance of people selected, “pulling names from a hat”

Convenience sample: Opposite of a random sample, only choosing people who are around you

1. 50 professional athletes from national sports teams, high-protein diet

2. 500 adults from various age groups, ethnic backgrounds, and geographic locations, daily exercise, mood levels

3. 150 adults from a gym in a wealthy urban neighborhood, stress reduction through exercise

1. Can’t be generalized because it was only high level athletes

2. yes, all the adults have different backgrounds

3. No, because all the adults are wealthy and go to the same gym

What does it mean to say when two things are correlated?

• Statistical way to measure the connection between 2 different things

• How well or does one factor predict the other?

Positive correlation

Negative corrrelation

No correlation

1. One goes up as the other goes up OR one goes down as the other goes down

2. One goes up as the other goes down, vice versa

3. No pattern at all

What is the “r” value?

• ALWAYS between -1 to 1

• -1 = perfect negative correlation

• 0 = no correlatinon

• 1 = perfect positive correlation

Why does correlation not equal causation?

There can be 3 explanations for a correlation:
1. One variable causes the second variable

2. The second variable causes the first variable

3. There is a 3rd variable which impacts the first and the second

Experimentation is needed to figure this out

Illusory correlations and Regression towards the mean:

Illusory correlations: The brain tries to find patterns that aren’t there. Just because two things are correlated doesn’t mean they are related

Regression towards the mean: People see great/poor results and forget that those are the extremes, not average. A person might see their favorite player have a great game, and the next game be disappointed when they see they do bad

What are experiments?

• Isolate cause and effect

• Manipulate one variable and see what the outcome is

Control vs Experimental group

Control group is the baseline group, no treatment, used to compare against the experimental group

Experimental group: The subjects are exposed to the treatment or conditions being studied

1. Random Assignment

2. Single Blind

3. Doube Blind

4. Placebo

5. Placebo Effect

6. Confounding variables

7. Dependent variable

8. Independent variable

9. Validity

1. Splitting a group into 2, experimental and control randomly. Important because it gets rid of any preexisting condition

2. Participiants do not know what group they are in, but experimenters do. Study is less likely to be biased

3. Neither the experimenters or participiants know what group they are in, helps remove even more bias, this time for the experimenter

4. A pill that has no treatment inside and is only used for psychological purposes. Used as a control to test against actual medicine

5. A condition/circumstance improving not because of the medicine, but because of your beliefs. Used to compare against the actual medicine

6. Variables outside the DV/IV that could skew the study. Important to get rid of so the study is legit.

7. Variable being measured/compared. Has a operational definition. Quantifies what is being observed

8. The variable that is being manipulated. This causes a change in the DV and what we want to observe.

9. Ensuring that the instruments used to collect data are accurate

Correlational studies vs experiments

Correlational studies establish connections between variables, but nothing more. Experiments on the other hand are often a result of correlational studies, do establish cause and effect. They do this by using a control and experimental group.