You are on page 1of 14

CORRELATION

By : Jubing 5
Definition Of Correlational
Research
• Gay in Sukardi (2008: 166) states that correlation research is
one part of ex-post facto research because in general
researchers do not manipulate the state of existing variables
and directly look for a relationship and the level of variable
relations expressed in the correlation coefficient.
• Correlation research is a study that involves data collection
activities to determine, is there a relationship and the level of
relationship between 2 or more variables.
• Correlation research is conducted, when the researcher wants
to know about the presence or absence and the strength or
weakness of a relationship of variables related to an object or
subject under study.
Characteristics of Correlational

• Correlation research is appropriate when complex


variables and researchers do not allow to manipulate
and control variables as in experimental research,
• Allows variables to be measured intensively in a real
setting or environment, and
• Enables researchers to obtain a significant degree of
association.
• (Sukardi, 2008: 166)
Types of Correlation
• Positive Correlation – when the value of one variable
increases with respect to another.
• Negative Correlation – when the value of one variable
decreases with respect to another.
• No Correlation – when there is no linear dependence or
no relation between the two variables.
Correlation Formula

Where, n = Quantity of Information


Σx = Total of the First Variable Value
Σy = Total of the Second Variable Value
Σxy = Sum of the Product of & Second Value
Σx2 = Sum of the Squares of the First Value
Σy2 = Sum of the Squares of the Second Value
Prosedur Dasar Penelitian Korelasional
• Problem Selection
Correlational studies can be designed to determine which variables from a list
of variables might be related, or to test hypotheses about an expected
relationship. Variables involved in research must be selected based on
inductive reasoning and deductive reasoning. In other words, the relationship
to be researched and investigated must be supported by theory or derived
from experience.

• Sample and Instrument Selection


Samples for correlational studies can be selected using an acceptable
sampling method, and 30 subjects are considered the minimum acceptable
sample size. In a study, it is important to choose and develop a reliable and
valid measurement of a variable to be examined.
NEXT...
• Design and Procedure
The basic correlational design is very simple; 2 or more scores obtained
from each number of samples selected, 1 score for each variable
studied, and paired scores are then correlated. The correlation
coefficient obtained indicates the degree or degree of relationship
between the two variables. Different studies investigate a number of
variables, and some use complex statistical procedures, but the basic
design remains the same in all correlational studies.

• Data Analysis and InterpretationI


the 2 variables are correlated, the result is the correlation coefficient. A
correlation coefficient in the form of decimal numbers, between 0.00
and + 1.00, or 0.00 and - 1.00, which indicates the degree or degree of
relationship between the 2 variables. When the coefficient approaches
+ 1.00; then the two variables have a positive relationship.
Kinds of Correlational
Studies
• Relationship Studies
Relationship studies are usually done in an effort to get an
understanding of any factors or variables related to complex
variables, such as academic learning outcomes, self-concept and
motivation. Variables that are known to have no relationship can
be eliminated from further attention or consideration.

• Prediction Study
Prediction studies are often done to make it easier to draw
conclusions about individuals or assist in the selection of
individuals. Prediction studies are also carried out to test
theoretical hypotheses about variables that are believed to be
predictors of a criterion, and to determine the predictive validity
of individual measurement instruments.
NEXT...

• Correlation and Causality


Correlational research is a study aimed at revealing
relationships between variables through the use of
correlational statistics (r). The square of the correlation
coefficient will produce the explained variance (r-square). A
correlational relationship between two variables is
sometimes the result of another source, so researchers
must be careful and correlation does not have to explain
cause and effect.
Correlational Research
Design
1. Bivariate Correlation
• The bivariate correlation research design is a research design that
has the objective to describe the relationship between two
variables. The relationship between the two variables is
measured. The relationship has a level and direction.

• The level of the relationship shows how or how strong the


relationship is, generally expressed in numbers between -1 and
+1, the level of the relationship is called the correlation
coefficient. The zero (0) symbol indicates there is no relationship
between variables. The correlation coefficient that moves toward
-1 or +1, is a perfect correlation at both extremes.
Next...
2. Regression and Prediction
If there is a correlation between 2 variables, and the researcher knows the
score on one of the variables, the researcher can predict the score on the
second variable. Regression refers to how well researchers can make
predictions of this kind. As the correlation coefficient approaches both -1 and
+1, researchers' predictions can be better. For example, there is a relationship
between health and stress.

3. Multiple Regression
Multiple regression is the expansion of regression and simple prediction by
adding several variables. The combination of these variables can give
researchers more power to make more accurate predictions. What the
researchers predict is called the criterion variable. What the researcher uses
to make predictions, while the variables that are known, are called predictor
variables.
Next...
4. Factor analysis
This one statistical procedure identifies the pattern of existing variables. A
large number of variables are correlated and the presence of high
correlations indicates a common important factor.

5. There are 2 designs that can be used to make statements about cause and
effect using the correlational method. The design is the path analysis design
and the cross-lagged panel design. Correlational Design Used to Draw Causal
Conclusions.

6. System Analysis
Systems analysis involves the use of complex or complex mathematical
procedures to determine dynamic processes, for example changes over time,
feedback straps, and flow and relationship elements.
Strengths and Weaknesses of
Correlational Research

• Correlational research has advantages including: its ability to


investigate the relationship between several variables together
(simultaneously); and correlational research is also able to provide
information about the degree of strength of the relationship
between the variables studied.
• this research is useful for overcoming problems related to the fields
of education, social, economics.

• Weaknesses of correlational research include: the results only


identify something in line with something, do not have to show a
causal mutual relationship; when compared to experimental
research, this correlational study is less orderly and strict, because it
lacks control over its independent variables; the interconnected
patterns are often erratic and vague or unclear; often stimulates its
users as a kind of short-gun approach, i.e. entering various data
without selecting and using any useful or meaningful interpretation.
THANK YOU

You might also like