Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
43 Cards in this Set
- Front
- Back
Nominal scale
|
- naming scale
- numbers have only the property of identity - mathematical operations: counting - statistical procedures: frequencies, chi-square |
|
Ordinal
|
- indicate rank ordering
- adding or multiplying, however, are meaningless (you cannot use a mean) - propertiesL identity, magnitude mathematical operations: less than or greater than statistical procedures: median, rank order, mann-whitney u test |
|
interval
|
- a scale that indicates an amount (but there is no 0 point)
- averages, SDs are meaningful but not ratios - properties: identity, magnitude, interval -mathematical operations:less than or greater than, addition or subtraction - statistical procedures - means, t-test, ANOVA |
|
Ratio
|
- a scale that has equal intervals with a true zero
- properties: many forms of measurement in physics are ratio, but this is rare in psych - properties: identity, magnitude, interval, true zero - mathematical operations: less than or equal to, addition or subtraction, multiplication or division - statistical procedures - means (ratios are meaningful), t-test, ANOVA |
|
Extraneous variable
|
Unplanned and uncontrolled factor(s) that can arise in an experiment and affect the outcome
|
|
Confounding variable
|
An unwanted factor that affect groups differently and make it difficult to know what caused changes in the dv.
- An extraneous variable can turn into a confounding varaible |
|
Observational Research
|
Recording naturally occurring behavior without intervention or manipulation of variables
|
|
Correlational Research
|
Identifying whether different behaviors, thoughts, and attitudes change together in predictable ways
|
|
Experimental Research
|
Manipulating the research setting to identify causes of behavior
|
|
Quasi-Experimental Research
|
Comparing existing groups (classification variable) to see if they differ
|
|
Qualitative Research
|
Studying people in their natural environment descriptively (without numerical analyses) and holistically
|
|
systematic observation
|
a form of observational research in which an investigator records behavior as it naurally occurs, attempting to note every behavior exactly as it emerges, often in a laboratory as an initial stage of research and prior to the development of hypotheses
|
|
naturalistic observation
|
a form of observational research involving the recording of behavior as it naturally occurs, without any attempt at intervention, often without the knowledge of those being observed
|
|
continuous real-time measurement
|
the measurement of the duration of behaviors as they occur
|
|
time point sampling
|
the measurement of the occurrence of a behavior by selecting specific points in time and recording whether the behaviors is occuring at that instant
|
|
time interval sampling
|
the measurement of behavior by noting whether it has occurred within a specified time interval or intervals
|
|
observer bias
|
the tendency on the part of observers to bring their biases and predispositions to the recording of data, a process that may be unintentional
|
|
behavior checklist
|
a list of behaviors to be recorded in naturalistic observation
|
|
observer drift
|
the tendency on the part of an observer to change criteria for recording behaviors over time
|
|
interobserver reliability
|
the degree to which two or more observers agree in their coding and scoring of behaviors they are observing and recording
|
|
measurement
|
assigning arbitrary symbols according to a predetermined set of rules to different events or objects.
|
|
overt behaviors
|
recording subject's behavior in terms of frequency, rate, speed, duration.
|
|
covert behaviors
|
self report- ask the people how they feel
|
|
law of small numbers
|
too much thought into small observations
|
|
illusory correlation
|
seeing a relationship where it does not exist
|
|
Avoid acquiescence
|
causes people to answer the same for every question
|
|
What’s necessary for Causation
|
- A precedes B
- A and B must covary (B must occur when A does) - A must be the most plausible cause for B with other potential causes ruled out |
|
truncated range
|
- If you don't have the full set of scores, it can be hard to see the relationship
- Ex. When you look at SAT scores you mostly look at people who got into college. If you looked at full range, you would see a pattern. |
|
Heterogeneous subsets/Multiple Populations
|
- Hard to detect
- Overall positive relation between two variables but each of the subsets has a negative correlation - When you combine subgroups into one overall data set, the correlation coefficient may not provide useful information |
|
Point-biserial
|
1 dichotomous variable, 1 continuous
|
|
Phi coefficient
|
2 dichotomous variables
|
|
Multivariate Analyses
|
- Simultaneously analyzing multiple variables
- Presented in a correlation matrix |
|
Moderator
|
specifies when the effects of a mediator will hold
|
|
Beneficence and nonmaleficence
|
Working for the benefit of people and avoiding harm to them
|
|
Fidelity and responsibility
|
Support for the discipline and working for the benefit of the community
|
|
Integrity
|
Honest application of psychology in research, teaching, and practice.
|
|
Justice
|
-Recognizing the implications one’s actions and striving to make good professional judgments
|
|
Guidelines for research
|
- Informed consent
- Is the pain and suffering reasonable? - Can the research be done without pain and suffering? - Can participants leave the experiment in a similar or superior state? - Does the value of the research outweigh the level of pain and suffering? - It is decided by a thierd, impartial party: IRB |
|
Random Sampling
|
Each individual in a population has an equal chance to be in the sample. This is simplest and best method. This is not the same thing as random assignment.
|
|
systematic sampling
|
variant on random sampling in which you sample every nth person.
|
|
stratified random sampling
|
population divided into subgroups & then these subgroups are randomly selected from using simple random sampling or systematic sampling. The size of the subgroups in the sample should equal the relative size of the subgroups in the population. Advantage is built in assurance that the sample will reflect the numerical composition of the various subgroups (only true if you know the numerical composition of the subgroups). Ex. Public opinion polling when you want the sample represents the population for gender, age, political affiliation, education, family income, etc.
|
|
cluster sampling
|
Rather than randomly picking some individuals from all subgroups of interest, you pick all individuals from some randomly chosen subgroups. Ex, picking classrooms for participation in a study, or sections of a particular class, cities, neighborhoods, or other naturally occurring clusters.
|
|
quota sampling
|
similar to stratified random sampling, except that after you identify the subgroups & the number of people you want, you get these people wherever you can find them.
|