Correlational & Quasi Experimental Designs

GROUP 4
ALTERNATIVE EXPERIMENTATION:
CORRELATIONAL
AND QUASI
EXPERIMENTAL
DESIGNS
REPORTERS:
SINILONG, ALAMA, VARGAS,

SAMANTHA C. ALEXANDRA E. ALYSSA NICOLE V.
OBJECTIVES
Learn more techniques that do not manipulate antecedent
conditions: correlations, other correlational-based
methods, and quasi-experimental designs

Learn how causal models can be constructed

from correlation-based designs

Understand how the results of these

nonexperimental techniques may (and may not)
be interpreted

HELLO EVERYONE,
LET'S PLAY A GAME
GUESS THE GIBBERISH

HOW TO PLAY?
In 'Gibberish Challenge' you basically needs to
guess a word or phrase.
The Game Master will show you a word that

makes no sense at all and you will need to
read it constantly to guess the right word.
YUH NET OFF
AH NULL LEI
SIZT
THE
ANSWER IS:
UNIT OF ANALYSIS
STUH TEST
TICK CALL ANA
LEI SIZT
THE
ANSWER IS:
STATISTICAL
ANALYSIS
LANE NAE YAR
RAY GRAY SEE
YAWN
THE
ANSWER IS:
LINEAR
REGRESSION
PA SEE TIVE
CALL REI LAY
SEE YAWN
THE
ANSWER IS:
POSITIVE
CORRELATION
CREWS SEC TIO
NULL STUH
DIES
THE
ANSWER IS:
CROSS SECTIONAL
STUDIES
CORRELATIONAL
DESIGN
CORRELATION
Degree of relationship, between two
traits, behaviors, or events.
CORRELATIONAL STUDIES
Is one that is designed to determine the correlation.
Often use correlational studies to explore behavior
that are not yet well understood.
SIMPLE CORRELATION
Relationships between pairs of scores from each subject are known as
simple correlations.
PEARSON CORRELATION
COEFFICIENT
The most commonly used procedure for calculating simple correlations
When r is computed, three general outcomes are possible;
a. positive relationship
b. negative relationship
c. no relationship
A. positive (direct)
relationship HYPOTHETICAL

RELATIONSHIPS
B. negative BETWEEN SIZE OF
(inverse) VOCABULARY
relationship AND LENGTH OF

DAILY TV
C. no strong VIEWING:
relationship
SCATTERPLOTS
Scattergraphs or scattergrams.
Visual representations of the scores
belonging to each subject in the study.
REGRESSION LINES
"Lines of best fit".
Illustrate the mathematical
equation that best
describes the linear
relationship between the
two measured scores.
POSITIVE inverse relationship
r= less than 0
CORRELATION
direct relationship NEGATIVE

r= +1.00 CORRELATION
COEFFICIENT OF DETERMINATION
Estimates the amount of
variability in scores on one
variable that can be
explained by the other
variable.
An estimate of the
strength of the
relationship between
them.
LINEAR REGRESSION
ANALYSIS
Used to describe data
and to explain the
relationship.
When two behaviors

are strongly related,
the researcher can
estimate a score on
one of the measured
behaviors from a
score on the other.
MULTIPLE CORRELATION AND
MULTIPLE REGRESSION
MULTIPLE MULTIPLE
CORRELATION REGRESSION
Represented by R, to test the relationship of Can be used to predict the score on

several predictor variables (X1, X2, X3,,,,) with one behaviors on the others.
a criterion variable (Y).
QUASI
EXPERIMENTAL
DESIGN
QUASI EXPERIMENTAL
DESIGNS
can seem like a real experiment, but they lack one or more of its
essential elements, such as manipulation of antecedents or
random assignment to treatment conditions.
can be used to explore or to investigate the same kinds of naturally

occurring events that we measure in correlational studies.
comparing different groups of subjects looking for differences

between them.
QUASI
Main Point:
We can never know for certain what
TREATMENT causes the effects we observe- so,

relative to true experiments, we say these
GROUPS
designs are low in internal validity.
Subjects:
The experimenter cannot exert
In the simplest quasi- control over who receives which
experiments, we form quasi- treatment because random
assignment is not possible.
treatment groups based on the
particular event, whose influence
Experiments vs. Quasi Experiments:
we want to investigate.
The amount of control the researcher

has over the subjects who receive
treatments.
EX POST FACTO STUDIES
A study in which the researcher
systematically examines the Ex post facto means
effects of subject characteristics
"after the fact."
(subject variable) but w/out
manipulating them.
IT DEALS WITH THINGS AS THEY OCCUR.
THERE IS NO MANIPULATION OF THE

CONDITIONS THAT INTEREST THE
RESEARCHERS.
INSTEAD OF STUDYING THE WHOLE RANGE

OF PEOPLE ALONG A PARTICULAR
DIMENSION; THE FOCUS CAN BE ON A
CAREFULLY CHOSEN SUBSET.
Next slide
It is more useful than certain kinds
It allows us to demonstrate of experiments, because the
that certain predictable information they provide helps us
relationships exist. understand the kinds of complex
behaviors that occur in real life.
The researcher
compares the effects of
different treatment
conditions on
NONEQUIVALENT preexisting groups of
participants
GROUPS
DESIGN The researcher cannot
exert control over who
gets each treatment
because random
assignment is not
possible.
IN TRUE EXPERIMENTS, OTHER OUTSIDE
INFLUENCES ARE CAREFULLY CONTROLLED, SO
THE ONLY SYSTEMATIC EFFECT ON BEHAVIOR
RESULTS FROM THE TREATMENTS, NOT FROM
ANYTHING ELSE.
WHEN USING THIS DESIGN, IT IS A GOOD IDEA

TO MEASURE SUBJECTS ON ANY ATTRIBUTES
THAT STRONGLY THREATEN INTERNAL
VALIDITY. THIS WAY, YOU MIGHT BE ABLE TO
DEMONSTRATE STATISTICALLY THAT THE
NONRANDOM GROUPS DID NOT DIFFER IN ANY
IMPORTANT WAY.
LONGITUDINAL
Often use surveys to collect
data that is either qualitative
DESIGN
or quantitative.
For a long-term study to be

successful, objectives and rules
are used in all areas of must be established from the
psychology, but they are beginning. As it is a long-term
particularly important for study, its authenticity is verified
psychologists studying human
in advance, which makes the
(and animal) growth and
results have a high level of
development.
validity.
Cross-sectional
studies capture a
a type of research specific moment
cross-sectional in time. National
design in which you
studies collect data censuses, for
collect data from
from many subjects instance, provide
many different
at a single point in a snapshot of
individuals at a
time. conditions in that
single point in time.
country at that
time.

CROSS SECTIONAL
STUDIES
WHY DO
THINGS
HAPPEN THE
WAY THEY DO?

QUALITATIVE VS.
QUANTITATIVE
Both based on Both capable if
epistemology of identifying causality

empiricism
Qualitative Research -
Quantitative Research -
scope if narrower but

scope is wider, but more
deeper shallow
VARIABLES
It is associated
A variable
A variable is a name
with
always contains
for something that is
thought to influence
quantitative
within it a
a particular state of
research and
notion of degree
being in something
with causal
or
else. models. differentiation.
VARIABLES
2 TYPES OF
VARIABLES
Independent (X) - cause
Dependent (Y) - effect
In the causal model, variation in
x is thought to cause variation in
y.
STATISTICAL
RELATIONSHIPS
1 POSITIVE 2 NEGATIVE
An increase in X causes an increase in

An increase in X causes a decrease in Y,
Y, or; or;
A decrease in X causes a decrease in Y. A decrease in X causes an increase in Y.
SPURIOUS
RELATIONSHIPS
A relationship is spurious
One indication of
when the observed

Identifying possible
spuriousness is that
relationship between X and

spurious relationships
Y is false. The observed

once you add another
relationship is actually
important for
variable, the
caused by a third variable. distinguishing

significance between
between correlation
the first two variables
Example: shoe size and

and causation.
reading
will disappear.
UNIT OF ANALYSIS
A unit of analysis is the entity under a study.
Example: individual, group, organization, nation
In the example of studying and grades, the
individual is the unit of analysis.

MODELING STATISTICAL RELATIONSHIPS
Association - positive
Independent Variable -
hours of study/week
Dependent Variable - grades
Unit of Analysis -
individual/student
MODELING STATISTICAL RELATIONSHIPS
Association -
Independent Variable -
Dependent Variable -
Unit of Analysis -
CAUSAL
CAUSAL
MODELING OR
RELATIONSHIPS
PATH ANALYSIS Is the manner in which a cause leads
Mathematical models
to its effects.
representing causal relationships
Cause - reason that something
within an individual system or

happens.
population. Effect - something happening.
TYPES OF CAUSAL
RELATIONSHIPS
CAUSAL CHAINS CAUSAL
HOMEOSTASIS
Is when one thing leads to another
thing, which leads to another thing,

Is when something supports its own
and so on. proliferation.

A➜B➜C A➜B➜C➜A
COMMON-CAUSE
RELATIONSHIPS COMMON-EFFECT
RELATIONSHIPS
Relationship is when one thing leads
Is when multiple things lead to one
to multiple things. thing.

A ➜ B and C A and B ➜ C
Example: Gabriel García Márquez uses
surrealist elements in a way that
illuminates personal and uniquely
Colombian insecurities about the past and

Causal Relationships
the future. That said, Márquez breaks the

in Your Thesis
boundaries of language and culture
because his unique stories are like fairy
tales—uncomfortable fantasies that strike
a chord at the level of the uncanny, where
“who and where” matters far less than
“how it feels.”
CROSS-
LAGGED PANEL
DESIGN
Is a type of
structural
equation model
that measures
two different
variables at two
points in time.
HOW TO ASSESS A CROSS-LAGGED
PANEL DESIGN
TWO SYNCHRONOUS
TWO SYNCHRONOUS
RELATIONS RELATIONS
HOW TO ASSESS A CROSS-LAGGED
PANEL DESIGN
TWO CROSS-LAGGED
RELATIONS
THANK
YOU

Correlational & Quasi Experimental Designs

Uploaded by

Copyright:

Available Formats

Correlational & Quasi Experimental Designs

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Correlational & Quasi Experimental Designs

Uploaded by

Copyright:

Available Formats

GROUP 4

SINILONG, ALAMA, VARGAS,

Learn how causal models can be constructed

Understand how the results of these

GUESS THE GIBBERISH

The Game Master will show you a word that

direct relationship NEGATIVE

When two behaviors

Represented by R, to test the relationship of Can be used to predict the score on

can be used to explore or to investigate the same kinds of naturally

comparing different groups of subjects looking for differences

We can never know for certain what

TREATMENT causes the effects we observe- so,

The amount of control the researcher

THERE IS NO MANIPULATION OF THE

INSTEAD OF STUDYING THE WHOLE RANGE

WHEN USING THIS DESIGN, IT IS A GOOD IDEA

For a long-term study to be

WAY THEY DO?

epistemology of identifying causality

scope if narrower but

for something that is

else. models. differentiation.

In the causal model, variation in

x is thought to cause variation in

An increase in X causes an increase in

when the observed

relationship between X and

Y is false. The observed

caused by a third variable. distinguishing

Example: shoe size and

In the example of studying and grades, the

individual is the unit of analysis.

within an individual system or

thing, which leads to another thing,

and so on. proliferation.

to multiple things. thing.

surrealist elements in a way that

illuminates personal and uniquely

Colombian insecurities about the past and

the future. That said, Márquez breaks the

because his unique stories are like fairy

tales—uncomfortable fantasies that strike

a chord at the level of the uncanny, where

“who and where” matters far less than

You might also like