Psychological Research: the basics

Hindsight Bias-- after learning the outcome of an event, you feel like you would have predicted that outcome

• common sense gives us insight to human nature-- tells us what has happened, not what will happen
• we are often more confident than we are correct
• over-confidence leads to overestimating our intuition
• scientific attitude needed: curiosity, skepticism, humility

Basic Research Vocabulary

1. theory= explanation that integrates principles and predicts behavior or events
2. hypothesis= testable prediction that enables us to accept, reject or revise the theory
3. operational definitions= statement of procedures used to define research variables-- concrete, unbiased way of measuring phenomena

Types of Psychological Research

• case study= observe one person for a long time (not a large sample, so often misleading)
• reveals underlying behavioral principles, tells us what can happen, suggests hypothesis for future
• survey= large amounts of quick, self reported data
• should be from a random sample of people, asking the right question can be really hard
• false consensus effect- tendency to over estimate the extent to which others share our beliefs and behaviors
• random sampling- each member of a certain population has equal chance of being in the study
• naturalistic observation= record behavior in natural situations without interfering or controlling

Correlation

• when traits accompany each other, they correlate
• correlation coefficient-- r= [-1,1] 0= no relationship, 1= strong relationship -1=strong inverse relationship
• correlation does not equal causation (depression and anxiety go together, but which one is the cause and which one is the result?)
• we must determine cause and effect through experimentation, not correlation
• illusory correlation= perception of a relationship where none exists (http://www.tylervigen.com/) 
• conformation bias= if we believe there is a relationship, we will notice and recall instances where our belief is confirmed
• we notice random coincidences and think that they are part of the bigger picture
• given random data, we look for orders and patterns
• correlational studies predict behaviors but don't explain cause and effect

Experimentation

• manipulate factors that interest us and keep other factors under control
• manipulated factors determine cause and effects
• dependent variable changes in response to independent variable
• placebo effect= results caused by expectation alone
• double-blind procedure= patients and experimenters should not be aware of which test group received the placebo so they don't dramatize data
• descriptive research- observes behavior
• correlational research- predicts behavior
• experimental research- explains behavior

Statistical Reasoning

• mode= most frequently occurring number
• mean= average number
• median= when lined up in descending/ascending order, the middle number

Measures of variation

• range= difference between highest/lowest numbers in the data
• standard deviation= how much the data varies around the mean number
• must differentiate between real difference and natural fluctuation
• observed difference is reliable if: deviations are small, sample is large and representative
• statistical significance= statistical statement of how likely it is that an obtained result occurred by chance alone
• alpha level 5%- chance/luck screw things up 5% of the time
• if it is significant, it is not necessarily important or practical