Comparing Research Strategies: Purpose

Descriptive: To describe individual variables as they already
exist within a specific group Correlational: To describe
the relationships between variables (does not attempt to
explain) Experimental: To explain a cause-and-effect
relationship between (usually) two variables

Comparing Research Strategies: Data

Descriptive: Measures variables of interest for every
individual Correlational: Measures two variables for each
individual and relates them to each other Experimental:
Creates two treatment conditions by changing the level of one
variable and measures 2nd variable for participants in each
condition (and relates groups to each other)

Experimental research: Steps

1. Manipulate the independent variable and observe dependent variable
to see if there are changes
2. To establish cause-and-effect, rule out possibility that changes
were caused by another variable

Experimental research: key terms

Independent Variable: Manipulated variable; usually two or more
treatment conditions Dependent Variable: Variable observed
for changes to assess affect of independent variable
Treatment Condition: A situation/environment characterized by
one specific value of manipulated variable Levels: Levels
of the independent variable; specific conditions used in the
experiment Extraneous Variables: Variables
other than independent or dependent that influence
the relationship between study variables Confounding
Variables: Variables that that vary with the changing level
of the independent variable

Types of extraneous variables

Environmental Variables Individual Differences
Time-Related Variables

When is an extraneous variable confouding?

Confound If it
influences the DV If it varies
systematically with the IV

Not a Confound (Not a threat) If
no influence on DV, then no
threat Example: individuals in a memory
experiment may wear different types of shoes (no threat)
A variable that changes randomly, without relation
to IV is no threat

Four elements of true experiments

Manipulation: Researcher manipulates one variable (IV) by
changing its value to create a set of two or more treatment
conditions, unique to experimental research Measurement: A
second variable (DV) is measured for a group of participants to
obtain a set of scores in each treatment condition.
Comparison: The scores in one treatment condition are compared
with the scores in another treatment condition. Consistent
differences are evidence that the manipulation caused changes.
Control: All other variables are controlled to be sure that they
do not influence the two variables being examined, unique to
experimental research

Manipulation

Steps for Manipulation Decide which values of
independent variable that you would like to examine
Create a series of treatment conditions corresponding to
those values Purpose of Manipulation
To allow researchers to determine the direction of a
relationship (crucial for experimental research)
Manipulation and Directionality Manipulation allows
researchers to observe directions of relationships
Manipulation and Third Variables Manipulation
helps researchers to control outside variables (instead of
waiting for changes to naturally occur) Manipulation
reduces influence of third variables

Manipulation check

Manipulation Check An additional measure to assess how
the participants perceived and interpreted the manipulation
and/or to assess the direct effect of the manipulation
How to Conduct Manipulation Checks Measure
independent variable to ensure changes occurred Ask
participants about the manipulation in a questionnaire

Why is manipulation check important?

Participant Manipulations: When manipulation involves something
within participant (such as frustration) Subtle
Manipulations: When variation is not salient and may not be noticed
by participants Simulations: When a study is simulating a
real-world situation and the effectiveness depends on participants�
perception and acceptance Placebo Controls: Placebos depend
on credibility so manipulation check assesses realism of the
placebo

Measurement

A second variable is measured for a group of participants to obtain a
set of scores in each treatment condition.

Comparison

Experiment Group: Treatment group in an experiment
Control Group: refers to the no-treatment condition in an
experiment Two types: no-treatment controls and placebo
controls

Comparison groups

No-Treatment Control Participants do not receive the
treatment being evaluated Placebo Control
Placebo � inert or innocuous medication that has no
medicinal effect Placebo Effect � a response by
participant even though placebo has no effect on the body;
person thinks medication is effective

Control

Definition Ensuring that the observed relationship is
not contaminated by the influence of other variables
Eliminating all confounding variables
Purpose of Control Experiments aim to show that the
manipulated variable is responsible for observed changes
Purpose is to rule out any other possible explanations for
observed changes
Control and Third Variables Very important to control
third variables that change along with the independent variable
and can affect the dependent variable Ways to Control
Extraneous Variables Holding Matching (or
counterbalancing) Randomization

Randomization

The use of a randomized process to avoid a systematic relationship
between variables

Random assignment

The use of random process to assign participants to treatment conditions

Controlling extraneous variables

Holding: An extraneous variable can be eliminated by holding it
constant or restricting it to a specific range Matching:
Values can be matched across conditions through matched
assignment Randomization: Use of random process to avoid
systematic relationship between two variables

Pro's and Con's of control methods

Holding Requires extra effort Used for 1 or 2
variables Limits generalization (and external
validity Matching Requires extra
effort Used for 1 or 2 variables
Randomizing Can control a wide variety of variables
simultaneously Not guaranteed 100% successful
(chance) Most commonly used

Elements of experimental research

Comparison: The scores in one treatment condition are compared
with the scores in another treatment condition. Consistent
differences are evidence that manipulation caused changes.
Control: All other variables are controlled to be sure that they
do not influence the two variables being examined.

External validity

External Validity: The extent to which research results can be
generalized to people, settings, times, measures, and
characteristics other than those used in the study Threat
to External Validity: Any characteristic of a study that limits the
ability to generalize the study�s results

Simulation

The creation of experimental conditions that simulate the
natural environment where the behaviors being studied would
naturally occur PURPOSE: to test �real-world� conditions in
a safe environment

Types of simulation

Mundane Realism: Superficial, usually physical, characteristics
of the simulation, which probably have little positive effect on
external validity Experimental Realism: Psychological
aspects of the simulation, extent to which participants become
immersed and behave normally, unmindful of the fact they are in an
experiment

Field studies

Research conducted in a place that the participant perceives is
a natural environment PURPOSE: to test real life behavior
in a natural environment

Pro's and Con's of simulation and field studies

Simulation PRO: Allows researchers to study life-like
situations CON: Researcher loses control and risks
threats to internal validity � study relies on participant
acceptance of simulation Field Studies
PRO: Allows researchers to study life-like situations
CON: Researcher loses control and risks threats to internal
validity � unpredictable participants