Lecture 4 Flashcards
(19 cards)
The Experiment
- Manipulation of IV and random assignment to levels of the IV
- Random assignment controls for confounding of potential individual difference variables
Quasi-Experimental Designs
Random assignment is absent and/or it is not possible to use random assignment. That is, the researcher has no control over the “independent” variable
Quasi-Experimental Designs: Nonequivalent Group, Post-test Only
The nonequivalent, post-test design consists of administering an outcome measure to two or more groups without random assignment
The Nonequivalent group, pretest-postest design partially:
eliminates a major limitation of the nonequivalent group post-test-only design by taking baseline measurements
What is a Naturalistic Study
A research method that involves observing and analyzing subjects in their natural habitat without manipulation or control by the researcher
What are the two components of a Raw Score?
True score + Error = Raw Score
What is a Raw score
the raw data from something like an intelligence test which may or may not show the person’s true ability
What are the effects of Error on the Raw Score
Error could increase or decrease the raw score
Two Types of Errors
- Random Error
- Bias
What is Random Error
Unsystematic or chance error
Ex. noise in the testing room
What is Bias
Systematic error
Ex. Wor with a tutor; poor instructors
What is the Floor Effect:
Scores gather at the low end of the measure (test is too difficult)
What is the ceiling effect:
Scores gather at the high end of the measure (test is too easy)
What is Reliability?
The extant to which a measure will produce the same results each time it is used
What will threaten Reliability:
Random Error
What is Validity
The extant to which a research method measures what it is intended to measure
What will Threaten Validity:
Bias (systematic error)
What is Probability Sampling
Where everyone in a population has an equal chance of being selected as part of the sample
What is Nonprobability Sampling:
Everyone does not have an equal chance of being sampled
