CORELATIONAL RESEARCH
Lecturer:
Prof. Dr. Moh. Muhassin, M.Hum
By:
Mohammad Gandi Pramudya
(1911040386)
FACULTY OF TARBIYAH AND TEACHER TRAINING
RADEN INTAN STATE ISLAMIC UNIVERSITY
LAMPUNG
2024 M/1445 H
PREFACE
Assalamualaikum Warahmatullahi Wabarakatuh,
The author praises and thanks Allah SWT. For all His gifts so that this paper can be
completed on time. The preparation of this document cannot be completed without the support of
many parties. the author would like to thank Mr Prof. Dr. Moh. Muhassin, M.Hum for the task
he can entrust. With this paper in mind, we can learn a lot from the delivery of the content in this
paper. Titled ‘correlational research’, it was prepared to complete the Research Methodology
assignment. In addition, this paper also aims to improve the understanding of the author and
readers in understanding one of the existing types of research.
Through this paper, the writer hopes that readers can gain knowledge and new
perspectives. This paper can also be used as a reference for higher education students who study
correlational research. After successfully making this paper, hopefully it can be useful for all of
us. The author realises that there are still some shortcomings in making this paper. Therefore,
constructive criticism and suggestions are expected for the improvement of this paper.
Bandar Lampung, November 17th, 2024
Mohammad Gandi Pramudya
CHAPTER I
INTRODUCTION
A. Background of the Problem
In the world of social and educational research, understanding the relationship between
two or more variables is often the foundation for predicting broader phenomena. One
effective method for analysing these relationships is correlational research. Correlational
research is a research approach that aims to identify and evaluate the relationship or
correlation between two or more variables, without manipulating or controlling these
variables. This research is widely applied in various fields, such as psychology, sociology,
education, and health sciences, because it allows researchers to explore relationships in
realistic and natural situations.
Correlational research provides important insights into the direction and strength of
relationships between variables. This research is often used to identify whether there is a
positive or negative relationship between certain variables. For example, in education,
researchers may be interested in understanding the relationship between learning motivation
and student academic achievement or the relationship between study habits and student stress
levels. By knowing these correlations, educators or educational psychologists can design
more effective programmes or interventions according to students' needs.
The main advantage of correlational research is its ability to utilise data from real
situations without having to intervene. This method allows researchers to collect data in a
natural context, so the results obtained tend to be relevant and reflect real conditions. In
addition, correlational research also allows researchers to reach variables that are difficult or
even impossible to manipulate in experiments, such as socioeconomic background or
personality characteristics.
However, although correlational research has many benefits, there is an important
limitation that needs to be considered, which is its inability to determine cause-and-effect
relationships. The relationship between variables found through correlation only suggests a
link, but cannot infer that one variable causes changes in another. Therefore, in practice,
correlational research is often followed by further research that is more experimental to test
whether the correlation found also contains a causal relationship.
Based on the important role of correlational research in various fields of study, this paper
will discuss the basic principles of correlational research, the types of correlations commonly
used, the advantages and limitations of this method, and its application in social and
educational research. By understanding the characteristics of correlational research, it is
hoped that researchers can be wiser in choosing the right method to answer their research
questions and interpret the correlation results more critically.
B. Formulation of the Problem
1. What are the basic principles in correlational research that researchers need to
understand?
2. How can correlation results be interpreted to determine the strength and direction of the
relationship between variables?
1
C. Objective of the Problem
1. To explain the basic principles of correlational research and the importance of
understanding this method in research.
2. To outline ways of interpreting correlation results, including the strength and direction of
the relationship between variables.
2
CHAPTER II
DISCUSSION
A. Definition Correlational Research
Correlational research is used to determine the extent to which two or more variables are
related. The level of relationship is indicated by the value of the correlation coefficient
which serves as a tool to compare the variability of the measurement results of these
variables.
Creswell argues that correlation research is research that provides an opportunity to
predict certain scores because of other scores and explains between variables. Based on the
above statement, there are two keywords in correlational research, namely correlation and
prediction. A correlation is a statistical test to determine the tendency or pattern for two (or
more) variables or two sets of data to vary consistently. A correlation is a statistical test to
determine the tendency or pattern for two (or more) variables or two sets of data to vary
consistently.
Gay in Sukardi states that ‘correlational research is a research study that involves
collecting data in order to determine whether and to what degree a relationship exists
between two or more quantifiable variables’. Correlational research is a study that involves
the act of collecting data in order to determine, whether there is a relationship between two
or more variables. If there is, what is the degree of relationship between two or more
variables. The existence of a relationship and the level of this variable is important, because
by knowing the level of relationship that exists, the researcher will be able to develop it in
accordance with the research objectives.
Correlation research, as Gay said, is one part of ex-postfacto research because usually
researchers do not manipulate the state of existing variables and directly look for the
existence of relationships and the level of variable relationships reflected in the correlation
coefficient. However, there are other researchers such as Nazir who classify correlation
research into descriptive research. On the other hand, according to Nazir, it is often treated
as descriptive research, because the research also seeks to describe current conditions in a
quantitative context reflected in variables.
In correlational research design, researchers use correlational statistical tests to describe
and measure the degree of association (or relationship) between 2 or more variables,
multiple sets of scores. In this design, researchers do not attempt to control or manipulate
variables as in an experiment. Instead they relate, using correlation statistics, two or more
scores for each person (for example, student motivation scores and student achievement
scores for each individual). Whereas correlation is a statistical test to determine the tendency
or pattern of two or more variables or two sets of data to vary consistently.
Correlational research is used when linking two or more variables to see if they influence
each other, such as the relationship between teachers supporting developmentally
appropriate practices and the use of a whole-language approach to reading instruction. This
3
design allows one to predict an outcome, such as the prediction that ability, school quality,
learning motivation, and academic task influence student achievement Anderson & Keith.
Correlational research can be easily understood when juxtaposed with causal research,
for example, experimental research. In experimental research, the researcher tries to
establish that one variable causes another, whereas in correlation, for example, the
researcher asks in the form of ‘what is the relationship between reading ability and
vocabulary acquisition? Not ‘does reading cause one's vocabulary acquisition to increase?’
B. Purpose of Correlational Research
The purpose of correlational research according to Arikunto is to find whether or not
there is a relationship and if so, how close the relationship is and whether or not the
relationship is meaningful. Specifically, the objectives of correlational research are: (1) to find
evidence of whether there is a relationship correlationbetween variables, (2) if there is a
relationship, to see the level of relationship between variables, and (3) to obtain clarity and
certainty whether the relationship is meaningful significant or insignificant.
C. Type of Correlational Research
Experts differ in classifying and categorising types of correlational research designs.
Shaughnessy and Zechmeiser in Emzir state that there are 5 types of correlational research
designs, namely a) bivariate correlation, b) regression correlation and prediction, c) multiple
regression, d) factor analysis, and e) correlation made to make causal inferences.
Meanwhile, Creswell (2015) states that there are only two main designs of correlational
research, namely explanatory and prediction.
1. Relationship Studies
Relationship studies are conducted in an attempt to gain an understanding of
complex factors or variables, such as academic learning outcomes, motivation, self-
concept. Variables that are known to be unrelated can be eliminated from further
consideration. Identification of related variables helps several key objectives. In
other words, the researcher tries to identify variables that are related to the dependent
variable and eliminate their influence that will not mix with the independent variable.
Relationship studies help researchers identify such variables to control them, and
then investigate the true influence of the variables.
In Creswell refers to relationship studies as explanatory correlational research or
‘relational’ research. Since one of the main purposes of this form of correlational
research is to explain the relationships between variables, we use the term
explanatory. An explanatory research design is a correlational design that interests
the researcher in the extent to which two (or more) variables covary, that is, changes
in one variable are reflected in changes in the other variable. Explanatory designs
consist of a simple relationship between two variables (e.g., sense of humour and
performance in drama) or more than two variables (e.g., peer pressure or feelings of
alienation contributing to binge drinking).
The specific characteristics of explanatory correlational research are as follows:
4
a. Researchers correlate two or more variables.
b. Researchers collect data at a single point in time.
c. The researcher analyses all participants as a single group.
d. The researcher obtains at least 2 scores for each individual in the group -one
for each variable.
e. Researchers report the use of correlational (or extended) statistical tests in data
analysis.
f. Finally, the researcher makes interpretations or draws conclusions from the
results of the statistical tests.
2. Prediction Studies
Two variables have a significant relationship, the score on one outcome variable
is used to predict the score on the other variable. The variable on which the
prediction is made is referred to as the predictor, and the predicted variable is
referred to as the criterion. Prediction studies are often conducted to facilitate
inferences about individuals or to assist in the selection of individuals. Prediction
studies are also conducted to test theoretical hypotheses about variables believed to
be predictors of a criterion, and to determine the predictive validity of instruments,
individual measurements.
In predictive design, researchers try to anticipate outcomes by using certain
variables as predictors. For example, supervisors and principals need to identify
teachers who will succeed in their schools. To select teachers who have a chance of
succeeding, administrators can identify predictors of success using correlational
research. Therefore, predictive studies are useful because they help anticipate or
predict future behaviour.
The intent of a predictive research design is to identify variables that will predict
an outcome or criterion. In this form of research the researcher identifies one or more
predictor variables and a criterion (outcome) variable. A predictor variable is a
variable that is used to make predictions about an outcome in correlational research.
In the case of predicting teacher success in school, the predictor might be
‘mentoring’ during teacher training or ‘years of teaching experience’. In many
predictive studies, researchers often use more than one predictor variable.
However, the predicted outcome in correlational research is called the criterion
variable. For example, teacher success is the criterion variable. Although more than
one outcome can be predicted, educational research usually includes only the
criterion variable.
The characteristics of correlational research prediction studies are as follows:
a. The author usually includes the word prediction in the title.
b. Researchers usually measure predictor variables at one point in time and
criterion variables at another point in time.
c. The authors predicted future performance.
D. Differences Between Relationship Studies and Predictive Studies
5
The main difference between relational research and this type of research lies in the
assumptions underlying the relationship between the variables under study. In relational
research, the researcher assumes that the relationship between the two variables is
bidirectional, in other words, he only wants to investigate whether the two variables have a
relationship, without having the assumption that one variable appears earlier than the other.
Therefore, both variables are usually measured at the same time. While in predictive
research, in addition to wanting to investigate the relationship between two variables, the
researcher also has the assumption that one of the variables appears before the other, or a
one-way relationship. Therefore, unlike relational research, the two variables are measured in
sequential time, that is, the predictor variable is measured before the criterion variable occurs
and cannot be the other way round.
E. Strength of Relationship Between Variables
According to Amos Neolaka (2016: 129) according to the theoretical study, the
interpretation of the strength of the relationship between two variables follows the guidelines
for interpreting the correlation coefficient as follows.
Coefficient Interval Relationship Level
0,00 No correlation
Very low
>0,00 – 0,199
Low
0,20 – 0,399
Medium
0,40 – 0,599
Strong
0,60 – 0,799
Very Strong
0,80 – 0,999
1,00 Perfect Correlation
When two variables are correlated the result is a correlation coefficient. A correlation
coefficient is a decimal number, between 0.00 and + 1.00, or - 0.00 and -1.00, which
indicates the degree of relationship between the two variables. If the coefficient is close to +
1.00; the two variables have a positive relationship. This means that a person with a high
score on one variable will have a high score on the other, and a person with a low score on
one variable will have a low score on the other; an increase in one variable is associated with
an increase in the other. If the correlation coefficient is close to 0.00, the two variables are
not related. This means that a person's score on one variable does not indicate the person's
score on the other variable. If the coefficient is close to -1.00, the two variables have the
opposite (negative) relationship. This means that a person with a high score on a variable will
have a low score on another variable; an increase in one will be associated with a decrease in
the other, and vice versa.
F. Steps in conducting correlational research
6
In correlational research, the steps used when planning or conducting a study according
to Creswell (2015: 702-708) are;
1. Determine whether correlational research best answers the research problem
Correlational research is used when there is a need to examine a problem that requires
identifying the direction and degree of relationship between 2 sets of scores. It is useful
for identifying the type of relationship, explaining the complex relationship of many
factors that explain an outcome, and predicting outcomes from one or more predictors.
Correlational research does not “prove” a relationship but shows the association or
relationship between two or more variables.
Because correlational research does not compare groups, it uses research questions,
not hypotheses. An example of a question in a correlational study is as follows:
a. Is creativity related to IQ test scores for elementary school children? (relates two
variables)
b. What are the factors that explain student teacher ethical behavior during the student
teaching experience? (explores a complex relationship)
c. Does high school class rank predict the GPA (Grade Point Average) of students in
their first semester in college? (prediction)
2. Identifying individuals to research
Ideally, individuals should be randomly selected from a larger population. The sample
should contain at least 30 individuals (N=30). Groups need to be of adequate size to be
able to use correlational statistics, e.g. N=30; larger sizes contribute to less error variance
and better claims of accuracy. A larger sample will increase the generalizability of the
results.
3. Identifying two or more measures for each individual in the study
The basic idea of correlational research is to compare participants in a group on two
or more specific characteristics, so the variable measures in the research question need to
be identified (e.g. literature search on previous research). The selection of the two
variables should be based on theories, assumptions, previous research results, or
experience that they are likely to be related. Next, instruments that measure these
variables need to be obtained. Ideally, the instruments should have proven validity and
reliability. Usually, a variable is measured on each instrument, but the instrument can
contain both variables correlated in the study.
According to Ibnu Hadjar (1996: 283) the instrument used must be able to reliably
measure the range or variability of subject responses. Some examples of the types of
instruments that can be used are tests, questionnaires, interviews, or observations.
4. Collect data and monitor potential threats
Lodico (2010:278) mentions that data for each variable must be obtained for each
person in the study. Since pairs of scores will be correlated, it is important to have a way
to link data from different measures so that scores that the same person has can be
grouped together. Next, describe the relationship between the two variables to predict the
outcome of the predictor variable. In Cresell (2015: 705) If, in a correlational study, the
data are taken (available) from the admissions office, the researcher does not have to
7
worry too much about procedures that threaten the validity of the scores. However, the
potential for a limited range of scores - little variation in the scores - certainly exists.
Other factors that might affect the researcher's ability to draw valid conclusions from the
results are the absence of standardized administration procedures, the conditions of the
testing situation, and participant expectations.
5. Analyze the data and present the results
The purpose of correlational research is to describe the degree of relationship
between two or more variables. Researchers look for a pattern of response and use
statistical procedures to determine the strength of the relationship as well as its direction.
A statistically significant relationship, if found, does not imply causation but merely a
relationship between variables.
Analysis begins with coding the data and transferring it from the instrument into a
computer file. After that the researcher needs to find the appropriate statistics to use.
Having found the most appropriate statistical test, the researcher then calculates whether
the statistic is significant based on the scores. In presenting the results, the correlational
researcher will present a correlation matrix of all variables as well as statistical tables.
6. Interpreting results
The final step in conducting correlational research is to interpret the results. This step
requires discussing the magnitude and direction of the results in correlational research, by
considering the impact of intervening variables in partial correlation research, by
interpreting the regression weights of variables in regression analysis, and by developing
predictive equations for use in predictive research.
In all of these steps, one of the overall concerns is whether your data supports the
theory, hypothesis or question. In addition, researchers consider whether the results
confirm or disconfirm findings from other studies.
G. Strengths and Weaknesses
According to Sukardi (2013: 170) the advantages and disadvantages of correlation
research.
Correlation research has advantages that can be explained as follows:
1. Useful in overcoming problems related to education, economics, and social fields,
because in this study it is possible for researchers to measure several variables and their
relationships simultaneously.
2. With correlation research, it is possible that several variables that have contributed to a
particular variable can be investigated intensively.
3. Correlation research generally conducts behavioral studies in realistic settings.
4. Researchers can perform predictive analysis without requiring a large sample.
While the weaknesses of correlation research that need to be considered by researchers are
that
1. With correlation research, researchers only identify what happens without manipulating
and controlling variables.
2. Researchers cannot establish a causal relationship.
8
9
CHAPTER III
CONCLUSION
From the discussion that has been presented above, we can conclude that:
1. The definition of correlational research is a study that involves the act of collecting data in
order to determine whether there is a relationship between two or more variables. If there is,
what is the degree of relationship between the two or more variables. The existence of a
relationship and the level of this variable is important, because by knowing the level of
relationship that exists, the researcher will be able to develop it in accordance with the
research objectives.
2. Types of correlation research designs are divided into two, namely
a. Relationship study
Relationship studies as explanatory correlational research or “relational” research. One of
the main objectives of this form of correlational research is to explain the relationship
between variables. The specific characteristics of explanatory correlational research are
Researchers correlate two or more variables, researchers collect data at one point in time,
researchers analyze all participants as a single group, researchers obtain at least 2 scores
for each individual in the group - one for each variable, researchers report the use of
correlational statistical tests (or their extensions) in data analysis, and finally, researchers
make interpretations or draw conclusions from statistical test results.
b. Predictive studies
The purpose of a predictive research design is to identify variables that will predict an
outcome or criterion. In this form of research, researchers identify one or more predictor
variables and a criterion (outcome) variable. A predictor variable is a variable used to
predict an outcome in correlational research. The predicted outcome in correlational
research is called the criterion variable. The characteristics of correlational research
prediction studies are that the author usually includes the word prediction in the title, the
researcher usually measures the predictor variables at one point in time and the criterion
variable at another point in time, the author predicts future performance.
3. The difference between relationship studies and predictive studies is that in relational
research (relationship studies) the researcher assumes that the relationship between the two
variables is bidirectional, in other words, he only wants to investigate whether the two
variables have a relationship, without having the assumption that one variable appears
earlier than the other. While in predictive research, in addition to wanting to investigate the
relationship between two variables, the researcher also has the assumption that one of the
variables appears before the other, or a one-way relationship.
4. The strength of the relationship between variables: When two variables are correlated the
result is a correlation coefficient. A correlation coefficient is a decimal number, between
0.00 and + 1.00, or - 0.00 and -1.00, which indicates the degree of relationship between the
two variables. If the coefficient is close to + 1.00; the two variables have a positive
relationship. If the correlation coefficient is close to 0.00, the two variables are not related.
10
5. Steps in conducting correlational research: (a) determine whether correlational research
best addresses the research problem,(b) identify individuals to study, (c) identify two or
more measures for each individual in the study, (d) collect data and monitor for potential
threats, (e) analyze data and present results, (f) interpret results.
6. Advantages in correlational research: (a) it is useful in overcoming problems related to
education, economics, and social fields, because in this study it is possible for researchers
to measure several variables and their relationships simultaneously, (b) with correlational
research, it is possible for several variables that have contributed to a particular variable to
be investigated intensively, (c) correlational research generally conducts a study of
behavior in a realistic setting, (d) correlation research is generally conducted in a realistic
setting, (e) researchers can perform predictive analysis without requiring a large sample.
7. Weaknesses in correlational research: (a) with correlational research, researchers only
identify what happens without manipulating and controlling variables, (b) researchers
cannot establish a causal relationship.
11
REFERENCE
Arikunto, Suharsimi. 2006. Prosedur Penelitain Suatu Pendekatan Praktik.
Jakarta : PT Rineka Cipta
Creswell, John. (2015). Riset Pendidikan Perencanaan, Pelaksanaan dan
Evaluasi Riset Kualitatif & Kuantitatif (Edisi Kelima). Yogyakarta:Pustaka
Pelajar.
Emzir. (2015). Metodologi Penelitian Pendidikan: Kuantitatif &
Kualitatif.Jakarta: RajaGrafindo Persada.
Hadjar, Ibnu. 1996. Dasar-Dasar Metodologi Penelitian Kwantitaif Dalam Pendidikan.
Jakarta : RajaGrafindo Persada.
Lodico, G, Marguerite, Dean T. Spaulding, Katherine H. Voegtle. 2006.
Methods in Educational Research From Theory to Practice. San
Fransisco. JosseyBass.
Neolaka, Amos. 2016. Metode Penelitian dan Statistik. Bandung : Remaja
Rosdakarya
Sukardi. (2013). Metodologi Penelitian Pendidikan Kompetensi dan Praktiknya.
Jakarta:PT Bumi Aksara.
12
