Stats Flashcards
Law of parsimony
The simplest explanation is often the best one
Empiricism
Studying things by collecting data
Case study
One individual is studied in great detail
Survey
Yields a lot of data about many different people
Representative sample
A sample representative of the population
Courtesy bais
People give socially acceptable answers
Naturalistic observation
Observing people/animals in their natural environments
Observer effect
People behave differently when watched
Participant observation
When researchers participate in a group in order to observe it without the observer effect
Observer bias
The observer’s opinion on what they expect to see interferes with their conclusion
Blind observers
People who observe a study without knowing the control/experimental groups
Correlation
Measure of the relationship between two or more variables
Correlation coefficient
The direction and strength of correlation (always between -1 and 1)
Experiment
Only method that allows researchers to determine cause
Operational definition
Clearly and indisputably defines how the variables of an experiment will be measured
Hawthorne effect
Behavior changes as a result of being part of a study
Confounding variables
Variables that interfere with each other (like how people studying more might change the results of that coffee experiment)
Placebo effect
Expectations interfere with behavior
Experimental effect
Same as observer bias but in an experiment
Single-blind study
Patients don’t know who’s the control group and who’s the experimental group
Double-blind study
Neither the patients or the researchers know who’s the control group and who’s the experimental group
Institutional Review Board (IRB)
Groups of psychologists and other professionals who make sure studies are ethical before they can be conducted
Common ethical guidelines:
Rights of participants must be weighed against the value of the study, participants must make an informed decision about their participation, any deception must be justified and followed by debriefing, participants can withdraw at any time, participants must be protected from risk, data must be confidential, and the researcher is responsible for helping any patients overcome negative consequences of the experiment
Informed consent
Parents give consent for their young children
Statistical analysis
Accounting for error in data
Descriptive statistics
Organizing numbers so they can be understood (included Measures of Central Tendency and Measures of Variability)
Frequency distribution
A table/graph that shows how numbers appear on a set
Histogram
Bar graph
Polygon
Line Graph
Normal curve
Bell curve
Skewed distribution
When the numbers are concentrated to the left (positive) or to the right (negative)
Bimodal distributions
When there are two number “peaks” in a graph
Measure of Central Tendency
Number that best represents the central part of frequency distribution (mean, median, and mode)
Standard deviation
Measures how useful the mean is. The smaller the better, ALWAYS positive
t-test
Measures if results are statistically significant, must be below 0.05 to be significant
Inferential statistics
Allows researchers to determine the difference between meaningful results and chance variations