Chapter 4 & 8* Flashcards
What are the two basic research designs?
non-experimental and experimental
Non-experimental design
relationships are studied by measuring or observing the variables of interest
Experimental design
involves direct manipulation of one variable, control of several other variables (those not of interest), then measurement of an outcome variable
Prediction
statement of the hypothesis that has been translated into the specific operationalizations of the study
Correlational studies
dominant type of non-experimental design; measure two variables in a group of people and determine if there is a relationship between the variables; used when one or more of the variables cannot be manipulated
Correlation coefficient (r)
a.k.a. Pearson product moment correlation; a statistic that indexes the linear relationship between two variables or expresses the strength of a relationship; ranges from -1 to 1
Correlation matrix
symmetrical table showing correlation coefficients between variables; used when there are many variables
What is the limit of a correlational design?
We can’t conclude causation
Internal validity`
ability to infer that independent variable causes changes in the dependent variable with alternative explanations being implausible
Two aspects of experimental design that allow causality
experimental control and random assignment
Independent variable
manipulated by the researcher; has multiple levels/conditions (usually control and experimental conditions)
Dependent variable
measured by the researcher; expected to be affected by the independent variable
What are the types of dependent variables?
self-report, physiological, and behavioral
Confounds or confounding variables
variable that is intertwined with and co-varies or correlates along with a variable of interest and can explain part or all of the result
What are the four most common relationships found in research?
(1) positive linear relationship (2) negative linear relationship (3) curvilinear or inverted U relationship (4) no relationship
Mediated relationship
describes how the relationship between two variables can be explained via a third variable
Mediating variable or mediator
a psychological (or physiological) process that occurs between an event and a behavioral response that helps to explain the relationship between two other variables
Construct
abstract variable, idea, or phenomenon
Random variability or error variability
a phenomenon that we can’t predict or explain, or that result from variables not of interest to us
Third variable problem
the possibility that a third, unmeasured variable (not intertwined) impacts both variables of interest
Experimental control
the only difference between conditions is the independent variable of interest
Random assignment
using chance to determine which participants end up in which conditions to control for the effects of extraneous variables not of interest
Field experiment
independent variable is manipulated in a natural setting out in the real world
What are two broad classes of experimental design?
between-subjects design (or independent groups design) and within-subjects design (or repeated measures design)
What are the three steps involved in any between-subjects experiment?
(1) obtaining two approx. equivalent groups of participants (2) introducing different levels of the independent variable to them (3) measuring the dependent variable
Selection differences
differences in the type of participants who make up each group in a between-subjects design
Matched pairs design or yoked design
Groups are made equivalent by first selecting pairs of participants who score the same on some variable of interest then randomly assigning one to the experimental group and the other to the control group
Solomon four-group design
the same experiment is conducted with and without the pretest then the researcher can examine whether is an interaction between the independent variable and the pretest variable (treated as another IV); used when there is concern for demand characteristics
Pretest-posttest design
the dependent variable is measured both before (pretest) and after (posttest) manipulation of the independent variable
Posttest-only design
the dependent variable is measured only once after manipulation of the independent variable
What are the three main reasons why a researcher may add a pretest?
(1) to counter problems associated with a small sample size (2) to select appropriate participants (3) when participants might drop out of the study
What is the advantage of a pretest?
it enables the researcher to assess whether groups were already roughly equivalent on some critical variable before the manipulation began
Selective attrition
a threat to internal validity when participant dropout results in a difference between conditions on some participant characteristic, causing a confound
What is the disadvantage of a pretest?
it can sensitize participants to what you are studying, enabling them to figure out your hypothesis
Within-subjects or repeated measures design
all participants experience all conditions and participants are measured on the dependent variable after being in each condition
What are the advantages of a within-subjects design?
fewer research participants are needed, it is extremely sensitive and able to detect differences between conditions
Order effect
the effect that the order of conditions has on the dependent variable in a within-subjects design
What are the three types of order effects?
practice effect, fatigue effect, and contrast effect
Practice effect
when performance improves because of repeated practice with a task
Fatigue effect
when participants perform worse over the course of a study because of diminished effort or fewer resources due to the passage of time
Contrast effect
when participants’ responses in a later condition are affected by a particular experience they had in an earlier condition
What are the two ways to deal with order effects in within-subject designs?
(1) using counterbalancing techniques (2) ensuring that the time between conditions is long enough to minimize the influence of the first condition on the second
Counterbalancing
a method of controlling for order effects in a within-subjects design by either including all possible orders of presentation for conditions or randomly determining the order for each participant; order is influencing both conditions equally
Latin square design (Partial Counterbalancing)
uses a limited set of all possible orders carefully constructed to (1) each condition appears first, second, third, and so on (2) and that each condition appears directly before and after each other condition exactly once
