Chapter 5—Experimental Research Flashcards
(11 cards)
What is an experiment?
A systematic research study where:
1. The researcher controls a factor(s)
2. The researcher hold all other factors constant
3. The researcher observes the results of the controlled variable’s variation
Describe Mill’s study; what is Agreement and Difference?
Mill discovered how, the methods of agreement and difference between variables together enable us to conclude with a high degree of confidence that X causes Y
Agreement: when Y exists following X; X is enough to produce Y. The experimental group satisfies this tenet
Difference: when X is absent, Y is absent; Y does not occur without X. The control group satisfies this tenet
What is a situational variable?
A varying feature in the environment that participants may encounter
- ex. # of bystanders influences the chance of someone helping a drowning dude
Describe a task variable
Varying the type of task participants are given to perform
What is an instructional variable?
Varying the instructions given to participants on how to perform a particular task
What is the difference between an experimental group and a control group?
An experimental group is those in the condition where the treatment is presented
A control group is those in the condition where treatment is withheld
What is the difference between extraneous variables and confounds?
An extraneous variable is a variable not of interest to the researcher; it may influence the DV if not controlled for
A confound is any uncontrolled extraneous variable that co-varies with the independent variable; it can provide an alternate explanation of the results
What are two main problems that can occur with a poorly chosen dependent variable?
Ceiling effect: when the DV is too easy or capped too low, so scores cluster at the top; if everyone scores near the maximum, you can’t find any differences
Floor effect: When the DV is too hard or has a minimum too high, so scores cluster at the bottom. Again, it’s hard to detect differences because everyone scores poorly
What are subject variables, and what is their unique position in research?
Pre-existing characteristics of the participants; they cannot be manipulated because they lack random assignment, so they have to select people for the conditions of the experiment based on the characteristics they already have
You cannot draw causation as a result, because without RA and manipulated variables…
Can only appear in between-subjects designs and mixed factorial designs
What is statistical conclusion validity?
The extent to which the researcher uses statistics properly and draws the appropriate conclusions from the statistical analysis
What are the 7 main threats to internal validity?
History: occurs when an event that could affect participants happens between the beginning of a study and its end, causing changes (i.e. work efficiency increase among rednecks could be due to Trump’s election)
Maturation: occurs when participants change from the beginning to the end of the study simply by maturing
Regression to the mean: if a subject records an extreme score on a pretest, the next measures will eventually cluster around their individual mean score.
Testing effect: the change in the DV might be due to repeated exposure (e.g., practice or sensitization), not the actual treatment or intervention.
Instrumentation effect: If the measurement tool or procedure changes between pre- and posttest, it becomes unclear whether the treatment caused the change or if it’s just due to the change in instrumentation.
Subject selection effect: If groups are not equivalent before the treatment starts, any observed differences might stem from those preexisting differences, not the IV. This is especially relevant in quasi-experimental designs.
Attrition: occurs if participants fail to complete a study; those remaining are different types of people than those who could not continue the study.
