PSYB04H3 Lecture Notes - Lecture 8: Control Variable, Internal Validity, Dependent And Independent Variables
PSYB04
LEC 8 - INTRODUCTION TO EXPERIMENTAL RESEARCH
March 2, 2017
Chapter 10
Experimental Variables
● Experiment : researchers manipulated at least one variable and measured another
○ Can take place in a lab and just about anywhere else
● Manipulated Variable: variable that is controlled
○ When researchers assign participants to a particular level (value) of the variable
● Measured Variable: records of behaviour or attitudes
○ Self reports, behavioural observations, or physiological measures
○ After experimental situation set up→ researchers record what happens
→ Independent and Dependent Variables
●
Independent variable →
manipulated variable
○ Independent variable’s levels are called conditions
○ Researcher has some “independence” in assigning ppl to diff levels of this
variable
●
Dependent Variable →
aka “outcome” variable
○ Measured variable
○ How a participant acts on the measured variable depends on the level of
independent variable
○ Researchers have less control over dependent variable
■ Manipulate the independent variable and observe what happens to ppl’s
self-reports, behaviours, or physiological responses
→ Control Variables
● When researchers manipulate an independent variable → need to make sure they’re
varying only one thing at a time
● Researchers also control potential third variables
○ Holding all other factors constant b/w the levels of the independent variable
○ Allow researchers to separate one potential cause from another → eliminate
alternative explanations for results
■ Important for establishing internal validity
●
Control variable →
any variable that experimenter holds constant on purpose
Why Experiments Support Causal Claims
● 3 rules for causation:
find more resources at oneclass.com
find more resources at oneclass.com
1.
Covariance →
is causal variable related to effect variable? Are distant levels of the independent
variable associated w different levels of the dependent variable?
2.
Temporal Precedence →
does the causal variable come before the effect variable in time?
3.
Internal validity →
Are there alternative explanations for the results?
→ Experiments Establish Covariance
● Independent variables answer “compared to what?”
○ Experiment is better source of info than your own experience
■ Allows you to ask the question: “compared to what?”
○ If independent variables didn’t vary → study cannot establish covariance
● Covariance: It’s also about the outcome
● Control groups, treatment groups, and comparison groups
○ Control group: level of an independent variable that’s intended to represent a
neutral condition
■ When study has control group, the other level(s) of the independent
variable called treatment groups(s)
■ Placebo group: when control group exposed to an inert treatment
● Example: a sugar pill instead of actual medication
○ Not all experiments need a control group
○ All experiments need comparison group → researchers can compare one
condition to another
■ Comparison group doesn’t need to be a control group
→ Experiments Establish Temporal Precedence
● Experimenters manipulate the independent variable to ensure that it came first in time
○ Advantage of experimental designs
→ Well-Designed Experiments Establish Internal Validity
● Study must ensure that the causal variable, and not other factors, is responsible for the
change in the effect variable
● Interrogate internal validity by exploring potential alternative explanations
● Confounds: alternative explanations
○ Potential threats to potential validity
○ Not sure what is causing change in the dependent variable
● Design confounds
○ Experimenter’s mistake in designing the independent variable
○ Systematically varies along with the intended independent variable → therefore
serves as alternative explanation for the results
● Selection Effects
○ Occurs in an experiment when the kinds of participant in one level of the
independent variable are systematically diff from those in the other
○ Can occur when experimenters let the participants choose which group they want
to be in
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Experiment : researchers manipulated at least one variable and measured another. Can take place in a lab and just about anywhere else. When researchers assign participants to a particular level (value) of the variable. Measured variable: records of behaviour or attitudes. Self reports, behavioural observations, or physiological measures. After experimental situation set up researchers record what happens. Researcher has some independence in assigning ppl to diff levels of this variable. How a participant acts on the measured variable depends on the level of independent variable. Researchers have less control over dependent variable. Manipulate the independent variable and observe what happens to ppl"s self-reports, behaviours, or physiological responses. When researchers manipulate an independent variable need to make sure they"re varying only one thing at a time. Holding all other factors constant b/w the levels of the independent variable. Allow researchers to separate one potential cause from another eliminate alternative explanations for results.