AP PSYCH Unit 0
Pseudoscience as a part of psychology: Humans coming with several cognitive biases that present roadblocks to critical thinking
Examples of biases allowing for pseudoscience: hindsight bias, overconfidence, confirmation bias, the tendency to perceive order in random events
Hindsight bias: (also called the “i knew it all along” phenomenon), the tendency to wrongly believe that one knew something before just being made aware of it. Like claiming they knew something beforehand when they find out by using false evidence
Overconfidence: the tendency to believe that we are capable, skilled, or correct than we actually are
Confirmation bias: the tendency to gather evidence that confirms preexisting expections; trying to find evidence for your own beliefs regardless of evidence proving it wrong
The tendency to perceive orders in random events: people will search for patterns even when there are none
The difference between pseudoscience and the actual scientific method: science disconfirms, rules things out, it can't deal with things that can't be falsified/tested, it predicts. Pseudoscience confirms and explains after the fact and pretends it's a prediction
The checks and balances of the scientific method: theory, hypothesis, operational definitions, replication, peer review
Theory: an explanation using an integrated set of principles that organizes and predicts observations
Hypothesis: a testable (falsifiable) prediction, often implied by a theory
How should hypotheses be phrased: positively in the future tense
Operational definition: description of how exactly measurements will be taken
Replication: repeating an earlier study to add to the evidence, but it wont work if the original study did not have clear operational definition
Peer review: experts in the field review each other's work before publication in academic journals
The 4 types of research methods: experiment, case study, naturalistic observation, correlation
Experiment: the only method that can allow for CAUSE AND EFFECT conclusions, requires a sample to make a conclusion about a population
All experiments involve: the manipulation of independent variables and random assignment into groups
Rule of the independent variable in an experiment: must be randomly assigned without self-select
Control group: the group that does not get the intervention being studied
Experimental group: the group that gets the intervention being studied (what was manipulated)
Variables of an experiment: independent variable, control group, experimental group, dependent variable
Random sample: a sample chosen from a population where everyone had an equal chance of inclusion; true random samples are nearly impossible
What the impossibility of true random sample leads to: convenience sampling
Convenience sampling: sampling people who are easily available to the researchers
Representative sample: a sample where the characteristics match the population. Consider: age, gender, socioeconomic status, region, and abilities
Sampling bias: a sampling method that produces an unrepresentative sample leading to potentially misleading results
Placebo: an inactive substance given to participants in studies with medications
Single-blind study: study where the participants don't know what they receive for the research; prevents participation bias (social desirability bias and placebo effect)
Social desirability bias: participants make answers to make themselves look good rather than telling the truth
Placebo effect: an inactive substance is given but the person claims to have felt better by believing it was some sort of medication to relieve something
Double blind study: neither the researchers nor the participants know what is being given; prevents experiment bias additionally
Effect of not using random assignment: introduces confounding variables
Confounding variables: factors other than the independent variable that might have influenced the findings
Case study: in-depth investigation of an individual or small group that may have a highly unusual trait
Benefit and drawback of case-studies: a lot of data, but likely not generalizable
Naturalistic observation: recording the natural behavior of many individuals, like parent-child interaction
Benefit and drawback of naturalistic observation: subjects can act differently when they know they are being observed. Doesn’t allow for control or cause an effect conclusions
Correlation: a technique used to identify relationships between two variables that can allow predictions to be made; it doesnt prove causation
Correlation coefficient r: a way to represent Correlation; the closer the absolute value of r is to 1, the stronger the relationship. -0.1 through -1.00 is the strength of a negative Correlation (opposite applying to positive)
Positive correlation: two things increasing or decreasing together
Negative correlation: one thing increasing while the other does the opposite
Problems associated with a correlation: third-variable problem and directionality problem
Third-variable problem: Third factor affecting the relationship
Directionality problem: there is no cause and effect shown by a correlation
Surveys: technique for ascertaining the self-reporting attitudes or behaviors of people
Problems associated with surveys: wording issues, self-report bias, social desirability bias
Wording issues in surveys: using the wrong wording to decrease approval of a survey, such as calling it “torture” instead of “enhanced interrogation”
Self-report bias: people can lie or misremember events, making them a wrong answer
Types of data: qualitative (narrative) data and quantitative (numerical data)
Ethical guidelines in experiments: informed consent, protection from harm, confidentiality, deception, debrief
Informed consent: participants understand procedures before agreeing to take part in research
Confidentiality: protecting private information shared by a client
Debrief: disclosure of deception, explaining the purpose of the experiment
Positive skew: skewed to the right, affects mean, so median should be used at times when finding average
Negative skew: skewed to the left
Normal curve in collecting data: when 68% of the population will fall within one SD of the mean, 95-96% of the population will fall within 2 SD; symmetrical
Example of normal curve: IQ mean is 100 and SD is 15, meaning 68% of the population would range from 85-115
Statistical significance: means that the results were likely caused by the effect the IV had on the DV, or that they were “not likely due to chance”
Effect size: inspired by statistical significance, how big the effect was when it comes to two comparisons in data (small 0.1-0.2-medium 0.3-07-large 0.8-1.0)