Chapter 10 Flashcards
Assuming something based constructs like intelligence or happiness. Constructs are intangible.
Inferences about constructs
Inferring from the data
Statistical inferences
Making claims about a causal relationship
Causal inferences
Assuming that the relationship between two variables can be generalized.
Inferences about generalizability
What does it mean when a generalizability inference is true.
If it is valid, then it can be combined with the three other inferences
whether the constructs (conceptual variables) that are being studied are the ones that are actually being manipulatied
construct validity
Whether the stats are sound and support statistical conclusions
statistical conclusion validity
A branch of psychology that studies research design, statistical analysis, and verification of results
Quantitative psychology
Inferential statistical tests require three assumptions
- a certain minimum number of observations in each cell
- that you are using a specific scale of measurement
- data will take the form of a normal distribution eventually
the degree to which we can be confident that the study demonstrated that one variable had a causal effect on another variable
Internal validity
The generalizability of the findings beyond the present study
external validity
External validity can be generalized to…
populations, settings and species
The process of repeating a study to determine if the original findings are consistant
replication
Why do we need replication?
one study cannot investigate everything nor can it make a study externally valid
Seven sources of threat to internal validity
History, maturation, testing, instrumentation, regression to the mean, attrition, selection
Referes to events that occur during a stay that are not part of the experiment that effect the experiment and threat internal validity.
History
The ways that people naturally change over time, independent of the study
maturation
how measuring a participants response might effect future responses
testing effects
Changes that occur to a measuring instrument that effect collection of data
instrumentation
The statistical concept that when two variables are not perfectly correlated, the more extreme scores will create lower correlation scores
Regression to the mean