| Terms |
Definitions |
|
Experiment
|
?
|
|
Generalisability
|
?
|
|
Correlation
|
?
|
|
Counterbalancing
|
?
|
|
Confounding variables
|
?
|
|
Correlation coefficient
|
?
|
|
Interviewer bias
|
?
|
|
Systematic sampling
|
?
|
|
Experimenter bias
|
?
|
|
Null hypothesis prediction
|
?
|
|
Randomization
|
bias distributed across groups
|
|
nominal
|
o Lowest scale of measuremento Variables whose vales differ by category (e.g. male/female)o Values of variable have different names but no ordering of values is impliedo You can cound numbers of observations falling into categories but cannot apply mathematical operations
|
|
prediction
|
specific statement concerning expected outcome of a study, particularly the relationship between dependent and independent variables
|
|
Attrition
|
drop out rates are different
|
|
Analogical Theory
|
• Relationship among variables explained through an analogy• One positive aspect: Provides good organizational schemeo If you already have something, and it is usually mechanical, you have something to help explain it o Allows us to collect datao Allows us to predict unexpected outcomes • Analogies are limited o Example• Brain as a computer• Computer doesn’t have emotions• It doesn’t form memory on its own
|
|
mode
|
the score that appears most frequently
|
|
Checking Data
|
look for invalid impossibilities, missing data, and outliers
|
|
participant observation
|
behavior observed in natural environment with researcher becoming part of group studied
|
|
outlier
|
an extreme score; typically scores that fall father than 3 standard deviations from the mean are considered outliers
|
|
SCIENTIFIC HYPOTHESIS
|
A formally stated expectaion about how a behavior operates
|
|
Measurement
|
The systematic assignment of numbers to objects or attributes of objects
|
|
closed ended questionnaires
|
1)used for inferential stats-quantitative methods
2)a multiple chocie exam
|
|
questionnaire
|
a type of survey that generates information about people directly by asking questions about experiences, attitudes, or opinions
|
|
Extraneous Variable
|
an uncontrolled variable other than the independent variable that may impact the dependent variable.
|
|
variable
|
anything that can be measured or manipulated
|
|
variables other than the independent variable that may bear any effect on the behavior of the subject being studied.
|
Extraneous variable
|
|
correlational research
|
research designed to examine the nature of the relationship between two measured variables
|
|
Descriptive
|
type of stats used to describe data
|
|
Respondent-driven sampling
|
Member of population recruits others for researcher (often with incentive like payment)
|
|
eta squared
|
effect size, amount of variance obatined
|
|
Interval scale
|
equal intervals, can “do math” with these - IQ and SAT scores, temp
"A scale with equal intervals but without a true zero point (e.g., temperature)"
|
|
recording single events versus behavior sequences
|
Traditional observational researchRecord single events in an observationBehavior sequences can also be recordedRecord behaviors occurring sequentiallyProProvides a more complex picture of behavior“Thicker” descriptionsConRequires more effort But . . . data yielded is worth the effort
|
|
stratified sampling
|
1)identify specific subgroups to be included in sample
2)then select equal random samples from each of the pre-identified subgroups
3)combine subgroup samples from each of the pre-identified subgroups.
Ex:Men & women in a study split population into different sexes and then randomly select from this sample.
|
|
Random Selections
|
increase the likelihood that the samp;le represnets the population
|
|
Mean can be used for which scales?
|
Interval
Ratio Scales
|
|
Rely on Theory
|
helps construct validity. good measures, and operational definitions.
|
|
block randomization
|
a procedure used to accomplish random assignment and ensure an equal number of participants in each condition, ensures that each condition of the study has a subject randomly assigned to it before any condition has a subject assigned to it again, also used in within subjects design as a counterbalancing procedure to ensure that when participants are tested in each condition more than once, they experience each condition once before experiencing any condition again
|
|
in case studies when soemone has to report something in the past, motivated forgetting in play
|
retrospective
|
|
biased assignment
|
a threat to internal validity that occurs when participants are assigned to conditions in a nonrandom manner, producing systematic differences among conditions prior to introduction of the independent variable
|
|
confound
|
something that causes a change in your dependent variable, but it isn't what we want to measure
|
|
confidence intervals
|
a range of acceptable values to determine the 95% confidence interval you calculate the boundaries M + or - 1.96 * o/{n
|
|
Absolute threshold
|
The average point on a sensory continuum (such as light intensity) at which an observer detects a stimulus
|
|
Quota Sample
|
a sample selected to include specified proportions of certain types of certain kinds of participants.
|
|
Experimental realism
|
How engaging is the simulation for participants
|
|
ANOVA (Analysis of Variance)
|
the most common inferential statistical tool for analyzing the results of experiments when dependent variables are measured on interval or ratio scales
|
|
Symmetric Transfer
|
a transfer that results from learning in an earlier condition, wherein A helps B or B helps A
|
|
Independent Variable (IV)
|
Variable that is manipulated or changed to test its effects on the dependent variable
|
|
Alternate Form Reliability
|
will scores correlate on two different forms of a test
|
|
social desirability bias
|
occurs when people respond to a question by trying to put themselves in a favorable light
|
|
nonequivalent groups posttest-only design
|
a quasi-experimental design in which two preexisting groups are studied - one that has received the quasi-independent variable and one that has not
|
|
Randomized Groups Design
|
each participant serves in only one condition of the experiment. (also known as between-groups/subjects design)
|
|
GOALS OF PSYCHOLOGICAL RESEARCH
|
Describe: each behavior and the conditions under which it occurs
Explain: behaivors in terms of their causes
Perdict: behaivors, know when it will occur or what will bring it about
Control: behaivors, we should be able to create the situation in which it occurs
|
|
Method of Authority
|
A method of acquiring knowledge in which a person relies on information or answers from an expert in the subject area.
|
|
Applied research
|
do research in a specific context, the research is geared toward a specific situation, usually asks for a specific group of people
|
|
odd-even reliability
|
equal number of correct answers on odd qestions as even ones
|
|
Inform participant beforehand of all aspects of research that might .....
|
influence willingness to participate
|
|
When do expectancy effects occur?
|
When experimenters make systematic errors in their interpretations of participants' performance or in their recording of date based on their expectations for participants performance
|
|
closed question
|
requires a yes or no, or answer by marking a point on some scale
|
|
experimental designs that control for validity
|
control selection of subjects, assignments to groups, and assignments of treatments:
|
|
Types of Bias in Experiments
Name two ways to minimize experimenter's role in experiment
|
Use other media to communicate instructions
2. Minimize experimenter/subject interactions
|
|
split-half reliability
|
people do better on first half of test then on the second half
|
|
Double Blind Procedure
|
to avoid bias - neither the subjects nor the researcher know who is in either group (a third party keeps records)
|
|
Weaknesses of within subject expts
|
Order and Sequence effects, people change after one test.
|
|
Power (of a statistical test)
|
occurs when a statistical test can detect effects.
most statistical tests are more sensitive to small effects with larger samples
The probability of rejecting the null hypothesis in a statistical test when it is in fact false
|
|
Institutional review board (IRB)
|
A board in nearly every United States institution conducting research that oversees the protection of human participants
|
|
What are the disadvantages of factorial designs?
|
1. # of participants may become inhibitive if there are many treatments2. can be difficult to interpret (particularly if interactions are significant)3. large commitment4. not designed to determine 'optimal' level of treatments
|
|
what are two threats to interal validty that occur naturally in treatments
|
history of the participants any changes in the participant that occur over time
|
|
Why is a one-group pretest-posttest design a bad idea?
|
None of the threats to internal validity are controlled (1x2)
|
|
APA Animal Research Guidelines
|
1. Must have a clear sci purpose - 2. must care for and house animals in a humane way - 3. must acquire animals legally - 4. muust deisgn experimental procedures that employ the lesat amount of suffering
|
|
What are two methods of probability sampling?
|
Need to have a sampling frame of people in population or need to use random-digit dialing (but not all members of the population may be included)
|
