Unit 1: Scientific Foundations of Psychology (Experimental Design and Statistics) Flashcards
Correlation
the measure of the extent to which 2 variables change together, and thus of how well either variable predicts the other.
Indicated by the correlation coefficient of two variables (r-value)
Illusory Correlation
the appearance of a relationship between 2 cariables when one does not actually exist.
Can happen when people only acknowledge confirming evidence
connection to confirmation bias
Correlation Coefficient
r-value
the statistical index of the correlation between 2 variables.
Correlation does not prove causation
Experimental Method
LIst basic steps
- Hypothesis
- Experiment design - establish an operational definition
- Pick a population/experimental group
- Conduct experiment: manipulate independent variable and measure effects on the dependent variable
- Debrief participants if necessary
Population
all those in a group being studied, from which samples are drawn.
Random Sampling
Randomly choosing members of a population so that each sample has an equal probability of being chosen.
Ensures that an experiment is generalizable to the public.
The sample fairly represents the population.
Sampling Bias
a flawed sampling process that produces an unrepresentative sample.
Hindsight Bias
the tendency to believe, after learning an outcome, that one would have forseen it (the i-knew-this-would-happen phenomenon)
Sample
Sampling is the process of selecting a representative group from the population under study.
A sample is said representative group
Experimental Group
group that recieves/is effected by the independent variable.
Control Group
all other variables are the same (at a base line), accept the independent variable
The standard to which comparisions with the experimental group are made.
Random Assignment
after random sampling experiments randomly assign people to the experiemental or control group to minimize the effect of preexisting differences.
minimizes the effect of confounding variables
Confounding Variable
any variable, aside from the IV that might affect the dependent variable
Random assignment helps minimize the effect og confounding variables
ex. age, time of day, amount of sleep participants recieve, culture, socioeconomic status
Experimenter bias
the expectations of the experimenter causes them to treat subjects in two groups differently, which affects results
Participant bias
participation knowing/believing they’re in the control or experimental group –> causing them to act differently
Double blind procedure
neither the participant nor the experimentor know who is in the control or the experimental group
mitigates participant/experimenter bias
Placebo Effect
peolpe who believe they recieving treatment (IV is being changed) will act/feel as it they did
Pro: able to assign causality
Con: limited by ethical concerns
Operational Definition
a carefully worded statement of the exact procedures used in a research study or how certain variables will be measured.
ex. intelligence is measured by numerical score on an IQ test
Allows experiments to be replicated
Case Study
a descriptive technique in which one individual/group is studied in depth over a long period of time in hopes of revealing universal principles.
Not very generalizable
Naturalistic Observation
obersinve and recording behavior in naturally occuring situations without trying to manipulate/control the situation.
Survey
a technique ascertaining the self-reported attitudes or behaviors of a particular group, usually by questioning a representative, random saple of the group.
Descriptive Statistics
used to represent/describe data from experiments
cannot be used to draw conclusions
examples:
- tables, diagrams, graphs
- measures of central tendency (mean, median, mode)
- variation - range, standard deviation
Inferential statistics
Help us analyze and draw conclusions from data
can be used to generalize results/draw conclusions
APA Guidelines for Humen Research
List all 6
- Informed consent: Participants must know that they are subjects, and must know the study entails
- Deception: allowed only if the study cannot be preformed without it
- Debriefing: subjects must be must be told the purpose/results of the study after it concludes
- Anonymity: researchers cannot disclose personal details of subjects without consent
- Coercion: cannot force people to take part via threats or offers they can’t refuse
- Risk: the risk cannot cause significant or long-term harm to the subjects
Statistical significance
How likely the results are to have occured due to random chance (instead of changes in the IV).
ex. p-value, t-test
Wording Effect
the possible effects on participants caused by the order of presented words or even the choice of the words themselves.