Chapter 4 Studying Behavior Variables Can have levels

Chapter 4 Studying Behavior • Variables – Can have levels – Types • 1. Situational – Characteristic of the environment • 2. Response – Reaction time, performance on a task • 3. Participant or subject – Gender, personality trait • 4. Mediating – Personal responsibility

Operational definition • A definition of the variable in terms of the operations used to measure it [the variable] • Defines a construct by specifying how it [the construct] is measured or manipulated • Converts an abstract concept into concrete, situation specific terms • For example – In animal research Hunger: deprivation for at least 12 hours – In human research • How hungry are you at this moment? a) Not at all b) Slightly c) Moderately d) very

Operational definition Necessary for empirical study Can be easily assigned or abstract Helps to communicate ideas in the scientific community However, researchers from different labs can define similar things differently

Variables • Usually thought to have a cause and effect connection • Independent variable – The “cause” – Manipulated in an experiment – Must have 2 or more levels • Can be quantitative (i. e. different doses of a drug) or qualitative (i. e. changing the sequence of words in a memory test). – It is usually on the horizontal axis (x axis) on a graph • Dependent variable – – The “effect” Measured in an experiment Can have one or more levels It is usually on the vertical axis (y axis) on a graph

For example • Homer notices that his shower is covered in a strange green slime. His friend Barney tells him that coconut juice will get rid of the green slime. Homer decides to check this out by spraying half of the shower with coconut juice. He sprays the other half of the shower with water. After 3 days of "treatment" there is no change in the appearance of the green slime on either side of the shower. • What was the initial observation? • Identify the – Control Group – Independent Variable – Dependent Variable • What should Homer's conclusion be?

Relationships b/w variables • Comparing values along a numeric scale – Positive linear relationship • Increases in one variable (on x axis) are accompanied by increases in another variable (y axis) • For example – Are fast talkers more persuasive? – Plot words spoken per minute (x axis) against attitude change (y axis) – Negative linear relationship • Increases in one variable are accompanied by decreases in the other variable • For example – Does the number of people in a group predict how efficient the group will be? – One study found that as the size of the group increased, the amount of noise decreased.

Relationships b/w variables • Curvilinear relationship – Increases in one variable are accompanied by decreases and increases in the other variable – Referred to as U shaped curves or inverted U shaped curves – For example • In dose response studies, in the case of a U shaped curve, – low doses can cause high responses – Intermediate doses can have low effectiveness – High doses can cause high responses – No relationship • For example – Does the length of the tail of a bird predict how many matings it will engage in?

Nonexperimental vs. experimental methods • Nonexperimental – Making observations of the variables of interest – Asking others to describe their behavior – Examining public records – Allows for observation of covariation between variables • Correlational method

Nonexperimental vs. experimental methods • Problems with correlational method – Direction of cause and effect – The 3 rd variable problem

Nonexperimental vs. experimental methods • Experimental – Involves direct manipulation and control of variables – The researcher manipulates the 1 st variable and observes the response – Reduces ambiguity in the interpretation of results • The direction of cause and effect is clear – Attempts to eliminate the influence of all potential confounding variables • Control of extraneous variables

Nonexperimental vs. experimental methods • Experimental control – Control of extraneous variables – Variables held constant cannot be a confounding variable – Accomplished by treating participants/subjects in all groups in the experiment identically • Randomization – Ensures that the extraneous variable is equally likely to affect all participants/subjects – Eliminates influences of individual characteristics – Ensures that the composition of participants/subjects in each group is identical

Causality • Inferences about causality require 3 elements 1. 2. 3. - • Is the cause “necessary and sufficient” for the effect to occur?

Choosing methods Field experiment • Done in the “field”, not the laboratory • The independent variable (i. v. ) is manipulated in a natural setting • Advantage: • Disadvantages: -

Evaluating Research • Validity – The accurate representation of information – Different types (each gives a different perspective on a research investigation) • Construct validity – The adequacy of the operational definition of variables – The measure has construct validity if it measures what it is supposed to – Face validity – Criterion oriented validity • Internal – The ability to draw conclusions about causal relationships from the data – Easier to prove when using the experimental method

Evaluating Research • Internal – Threats » Biased assignment of subjects » Experimental confounds (i. e. performing systematic differences) » Differential attrition » Pretest sensitization • External – The extent to which the results can be generalized to other populations and settings – Can these results be replicated? With different participants? • No method is superior to another