Research Methods (8-10%) Flashcards
Critical thinking
the ability and willingness to assess claims critically and make objective and supported arguments
Psychological Test
procedures used to measure and evaluate personality traits, emotional states, and values through oral and written tests
Valid
A test measures what it is supposed to measure accurately
Reliable
measures the same every time
- if it is valid it is automatically reliable
Experimental method
hypothesis - prediction of how multiple factors are likely related
variables - factors that can have different values (affects results)
experiment - researcher manipulates variable under controlled conditions and observes response
Independent variable
The experimental factor that is manipulated; the variable whose effect is being studied.
dependent variable
variable that changes as a result of a change in the independent variable
Cause and Effect
A controlled experiment can only establish…
Population
all individuals in the group of study
Sample
portion of the population that is chosen for experiment
Random Selection
all people in the population have an equal chance to be chosen
Random Assignment
All the selected have equal chance to be chosen for experiment or control
Experimental Group
receives the treatment
Control Group
doesn’t receive the treatment
Norms
establish standards of performance
Confounding Variables
anything other than the independent variable that is different between the groups
Operational Definition
describes the specific procedure used to determine the presence of a variable
- must be precise for the experiment to be repeatable
Experimental Bias
researchers expectations or performances influence the results
Single Blind Procedure/Experiment
Participants don’t know if they are control or treatment group but the experimenter does
Double Blind Procedure/Experiment
experimenter and participant don’t know if they’re control or treatment group
- gets rid of confounding variables
Placebo
A false treatment
Placebo Effect
subjects sometimes believe that the treatment will be effective, and they then experience an improvement in wealth or well being
- mental game
Quasi Experiment
experiment on a specific group of people before actual experiment to narrow it down
- random selection is not possible
Correlational Methods
look at the relationship between two variables without establishing cause and effect relationships
- the goal is to determine to what extent one variable predicts the other
Correlation Coefficient
varies from 1 to -1
Direct Positive Correlation
1
- one variable changes and the other variable changes also
Direct Negative Correlation
-1
- one thing goes up the other goes down
No Relation/Correlation
0
-One variable will not predict the other
Naturalistic Observation or Behaviorist
- carried out in the natural environment
- person being researched cannot know they are being researched on
Hawthorne Effect
people will alter their behavior because of the researchers attention and not because of the actual treatment
Survey
questions or interviews to ask large numbers of people
advantage- cheap and reaches a lot of people
disadvantage- people cheat or lie
Random Sampling
participants are representation of population
Ex Post Facto
starts with effect and seeks cause
Case Study
method of research that involves an in-depth examination of a specific group or single person that typically includes interviews observation and test scores
- can be done when you can’t do experiment because it is unethical
Descriptive Statistics
numbers that summarize a set of research data obtained from a sample
Central Tendency
describes the average most typical scores for a set of research data or distribution
- what is normal
Mode
the most frequently occurring score
Bimodal
two or more frequents
Multimodal
three or more frequents
Median
the middle score when data is organized by size
Mean
average of a set of scores
Range
the top score minus the bottom score
P-value/score
likelihood or result is caused by chances
-p is greater than or equal to .05 ( can be no greater than 5%)
-if value is higher than 5% it is not valid
P
probable
Positive Skew
mean is larger than median
Standard Deviation
unit of measurement to tell us how far from normal
Z-Score
your number of standard deviations away from the mean
Negatively Skewed
distribution is squeezed into the negative end of distribution
- mean is lower than the median
Positively Skewed
skewed towards the positive side of distribution
- the mean is greater than the median
Negative Skew
the mean is smaller than the median
Normal Distribution
percentages pic
