Ex Post Facto Research

Ex post facto research is also appropriate when the variable actually could be
manipulated but is not because it would be unethical or irresponsible to do so. For example, it
would not be ethical to manipulate illegal drug use or use of alcohol or cigarettes to study
their effects on human subjects. To study the effect of retention on subsequent achievement,
you would not want to randomly assign some children to be retained and others to be
promoted.

The designation ex post facto, from Latin for “after the fact,” indicates that ex post
facto research is conducted after variation in the variable of interest has already been
determined in the natural course of events. This method is sometimes called causal
comparative because its purpose is to investigate cause-and-effect relationships between
independent and dependent variables.1 Ex post facto research, unlike experimental research, does
not provide the safeguards that are necessary for making strong inferences about causal relationships.
Mistakenly attributing causation based on a relationship between two variables is called the post hoc
fallacy.Ex-post facto research is a study in which independent variables have occurred when
researchers began by observing the dependent variables in a study. ex post facto research was
an empirical discovery carried out systematically, researchers did not exercise control over
independent variables because their manifestations had occurred 2. The name ex post facto
itself in Latin means "from after facts". This shows that the research was carried out after the
differences in the independent variable occurred because of the development of the incident
naturally3.ex post facto research was an empirical discovery carried out systematically,
researchers did not exercise control over independent variables because their manifestations
had occurred

Ex-post facto research aims to find the causes that allow changes in behavior, symptoms or
phenomena caused by an event, behavior, symptoms or phenomena caused by an event, behavior or
things that cause changes in the overall independent variable have occurred. Based on the description
above it can be concluded that Ex-Post Facto research is a study that explains or discovers how the
variables in the study are related or influential, and find out how the symptoms or behavior occur.

1
Ary,D.,Jacobs,L.C., Sorensen,C.,and Ravazier,A. Introduction to Research in Education.(Boston: Wadsworth
Cengange Learning Products.2010)page 332
2
Ary, Donal. (1982). Pengantar Penelitian Dalam Kependidikan. Surabaya : Usaha Nasional.
3
Darmadi, Hamid. 2011. Metode Penelitian Pendidikan. Bandung: Alfabeta
 For example, a researcher wants to know the effect of smoking on the ability to absorb
oxygen in the blood. Researchers were not likely to carry out experiments by telling people to smoke
several cigarettes a day to find out their influence on the ability of blood to bind oxygen.

B. Planning an Ex Post Facto Research Study

1.) The first step in an ex post facto study is to state the research problem, usually in the form
of a question.

For example :

What is the relationship between variable A and variable B? Or, what is the effect of variable
A on variable B?

in this type of research the variables involved are those that the researcher does not directly
manipulate. The researcher then states a hypothesis about the expected relationship and defi
nes the variables in operational terms.

2.)Next, select two or more groups to be compared. Recall that investigators doing ex post
facto research achieve the variation they want not by directly manipulating the variable but
by selecting individuals in whom the variable is present or absent, strong or weak. Thus,
these two groups should differ on the variable of interest in that one group should possess the
characteristic and the other group should not, but they should be similar on any relevant
extraneous variables. Differential (subject) selections threat to the internal validity of ex post
facto investigations because you have no control over the selection of subjects into the two
groups. They are selected because they already possess the variable of interest

For Example :

3. Determine whether your question requires a proactive or a retroactive design. There are
two types of ex post facto research designs—proactive and retroactive

a. The proactive ex post facto research design begins with subjects grouped on the basis of an
independent variable. such as father present/ father not present or retained/promoted. The
researcher then compares these preexisting groups on measurers of dependent variables such
as self-confi dence, mental health, and academic performance.

b. Retroactive ex post facto research seeks possible antecedent causes (independent

variables) for a preexisting dependent variable.
for example :

Whether students graduate from high school or drop out is a variable that cannot be
manipulated. Therefore, a researcher would use retroactive ex post facto research to
investigate hypotheses about possible causes, such as truancy, attitude toward school,
ambition, and parents’ education.

C. Ex-Post Facto Research Types

1. Correlation Research

Correlation research is a study that involves the act of collecting data to determine whether
there is a relationship and the level 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 the relationship that exists, researchers will be able to develop it according to the
research objectives. 4Correlation research aims to find whether there is a relationship and if
there is, how close the relationship is and means or not the relationship 5. Meanwhile Ricards,
Platt and Weber provide a definition of correlation as a measure of the strength of the
relationship between two data sets. This method describes quantitatively the association or
relation of one variable with other variables. 6.Based on the definition above, the keyword of
correlational research is "the relationship between variables." That is, this study tries to find
out the extent of the relationship and the quality of the relationship.

Example of Correlational Research is we want to know between the value of a math test
group of students with high test scores

2. Comparative causal research (ex post facto) is included in the quantitative research
method. Quantitative research is a process of finding knowledge that uses data in the form of
numbers as a means of finding information about what we want to know and used to research
certain populations or samples, data collection using research instruments, statistical data
analysis, with the aim of testing hypotheses set.7

4
Sukardi, Metodologi Penelitian Pendidikan: Kompetensi dan Praktiknya,page 166.

5
Suharsimi Arikunto, Prosedur Penelitian: Suatu Pendekatan Praktik, (Jakarta: PT. Rineka Cipta, 2013), page 313.
6
Richards, Jack, John Platt, and Heidi Weber, Longman Dictionary Of Applied Linguistics, (Harlow, Essex, England:
Longman, . 1985), h. 66
7
Wicaksono, Andri. 2014. Metode Penelitian Kausal Komparatif Ex Post Facto.
based on the explanation about Comparative casual research can be concluded comparative
causal research is research that is directed to investigate causal relationships based on
observations of the consequences that occur and look for factors that are the cause through
the data collected. In this study the basic approach is to begin with the difference between
two groups and then look for factors that might be the cause or effect of these differences.

The purpose of this study is to investigate the possibility of a causal relationship based on
observing the consequences, and looking for facts that might be caused by certain data.
Causal-comparative research has characteristics that are ex post facto, meaning that data is
collected after all the events in question take place (past). Research takes one or more
consequences (as "dependent variables") and tests the data by tracing back to the past to look
for causes, mutual relations and their meanings and tends to rely on quantitative data.

Comparative causal research (ex post facto) is carried out in five stages, such as:
1.) Determination of research problems

In the formulation of research problems or research questions, we speculate about the causes
of phenomena based on previous research, theory, or observation.

2.) Determination of groups that have the characteristics to be examined

For example students who have been able to use a computer before entering elementary
school, because in their home there is a computer can be seen by looking at the smallest
homogeneous group that has the critical variable.

3.) Selection of comparison groups

Taking into account the characteristics or experiences that distinguish groups must be clear
and defined operationally (each group represents a different population).

4.) Data collection

Data collection can be done using research instruments that meet the requirements of validity
and reliability.

5.) Data analysis

Data analysis begins with descriptive statistical analysis calculating the average and standard
deviation. Furthermore, a more in-depth analysis is conducted with inferential analysis.
For example :

A lecturer in one faculty requires first-level students majoring in Indonesian in front of his
friends. It was discovered that there were those who were fluent and those who did not,
especially in using Indonesian, even though they were even though students majoring in
Indonesian. The title "Effect of Mother Language, Outdoor Environment, and Indonesian
Language Learning in High School on the Proficiency of Speech Level I Students of the
Indonesian Language Department" can be used.8

1. Identify the problem:

Research assumes that there is a causal relationship between the three factors in the title
above to the ability to speak. Indonesian language lessons in high school affect proficiency.

2. Independent variable:

a. Mother tongue
b. Environment outside the home
c. Indonesian language learning in high school
3.) Dependent variable: Speech proficiency

4.) Problem Formulation:

Are the factors of mother tongue, the environment outside the home, and Indonesian
language lessons in high school affect the proficiency of speech.

5.) Hypothesis:

The factors of mother tongue, the environment outside the home, and Indonesian language
lessons in high school significantly influence the proficiency of speaking in Indonesian
language majors.

A weakness of causal–comparative studies is that, because the cause under study has
already occurred, the researcher has no control over it. For example, suppose a researcher
wanted to investigate the effect of heavy smoking on lung cancer and designs a study
comparing the frequency of lung cancer diagnoses in two groups, long-time smokers and
nonsmokers. Because the groups are preexisting, the researcher did not control the conditions
under which the participants smoked or did not smoke (this lack of researcher control is why
the variable is known as a grouping variable, rather than an independent variable). Perhaps a
8
Damayanti, Deni. 2013. Panduan Lengkap Menyususn Proposal, Skripsi, Tesis, Disertasi untuk Semua
Program Studi. Yogyakarta: Penerbit Araska.
large number of the long-time smokers lived in a smoggy, urban environment, whereas only a
few of the nonsmokers were exposed to those conditions. In that case, attempts to draw
cause–effect conclusions in the study would be tenuous and tentative at best. Is it smoking
that causes higher rates of lung cancer? Is it living in a smoggy, urban environment? 9

D. STEPS FOR IMPLEMENTING EX POST FACTO RESEARCH

1. Formulation of the problem, the problem set must contain the cause for the emergence of
the dependent variable, which can be known based on the results of research that has been
done or the interpretation of researchers towards the results of observations of the
phenomenon being studied

The problem of this research can be in the form of a hypothesis statement or purpose.
Hypothesis formulation is used if the nature of the difference can be predicted by the
researcher before the data is collected. Whereas the purpose statement is used if the
researcher cannot predict differences between subject groups that are compared in certain
variables

2.) After the problem is formulated, the researcher must be able to identify comparative
hypotheses or alternatives that might explain the relationship between independent and
dependent variables.

3.) Determination of the group of subjects to be compared. First of all, the selected group
must have the characteristics of the research focus. Furthermore, the researcher selects groups
that do not have these characteristics or differ in their ranks

4.) Data collection. Only the required data is collected, whether it is relevant to the dependent
variable or with respect to the factors that are possible to emerge a rival hypothesis. Because
this research investigates phenomena that have already occurred, often the required data is
available so that it is researched to select the appropriate source. Besides that various
instruments such as Les, questionnaires, interviews, can be used to collect data for research

5.) Data analysis. The data analysis technique used is similar to that used in differential and
experimental research, where the comparison of variable independent values is carried out
between subject groups on the basis of the concentration factor. This can be done by T-test

9
Gay,I.R. Geoffrey ,E. Mills, Peter,W. educational research-competence for analysis and applications(New
Jersey: Pearson Education.2012)page 19-20
analysis techniques, independent or ANAVA, depending on the number of groups that is.
Whatever the statistical analysis techniques

6.) inferential used, usually the analysis begins with the calculation of the mean or mean and
standard deviation to find out the inter-group comparison descriptively.

7. Interpretation of basil. The cause and effect statement in this study needs to be done
carefully. The quality of relationships between independent and dependent variables is very
dependent on the ability of researchers to choose a homogeneous comparison group and the
belief that the emergence of counter-hypotheses can be prevented.10

E. PARTIAL CONTROL IN EX POST FACTO RESEARCH

There are strategies for improving the credibility of ex post facto research, These strategies
provide partial controlof the internal validity problems of common cause and other possible
independent variables. Among these strategies are matching, homogeneous groups, building
extraneous variables into the design, analysis of covariance, and partial correlation. 11

a. Matching

A common method of providing partial control in ex post facto investigations is to match the
subjects in the experimental and control groups on as many extraneous variables as possible.
This matchingis usually done on a subject-to-subject basis to form matched pairs. First, using
matching in an ex post facto study assumes that you know what the relevant factors are—that
is, the factors that may have some correlation with the dependent variable. Even if you are
able to match on a few relevant variables, you will leave many other variables unmatched,
and these unmatched variables may affect the dependent variable. Furthermore, matching is
likely to greatly reduce the number of subjects that can actually be used in the final analysis.
The loss of cases inherent in the matching process is an even more serious problem in ex post
facto research than in experimentation, in which matching precedes the measurement of the
independent variable.

A more serious problem than loss of subjects is the role of regression in an ex post facto
matched-pairs design. Matching looks good because it provides experimental and control

10
Handayani,M.et all. Konsep Teori dan contoh desain penelitian Ex Post Facto.(Jakarta:Universitas Indraprasta
Pgri.2015) page 6-7

11
Ary,D.,Jacobs,L.C., Sorensen,C.,and Ravazier,A. Introduction to Research in Education.(Boston: Wadsworth Cengange
Learning Products.2010)page 337-342
groups that are equal on a pretreatment variable or variables. However, when two matched
groups are drawn from different populations, regression toward the original population means
will be expected to create spurious results whenever the two populations are not equal.
Matching pairs from within a singlepopulation is often a useful strategy. Matching subjects
from one population with subjects from another population is a badstrategy. As is the case
with change scores, matching only partly adjusts for preexisting differences between groups,
and this underadjustment can be misleading.

b. Homogeneous Groub

You may recall from the discussion of control in experimentation that it is possible to control
for the effects of a variable by selecting samples that are as homogeneous as possible on that
variable. A similar procedure can be followed in ex post facto research. Instead of taking a
heterogeneous sample and comparing matched subgroups within it, an investigator may
control a variable by including in the sample only subjects who are homogeneous on that
variable. If academic aptitude is a relevant extraneous variable, the investigator could control
its effect by using controlled, you can use only males or only females in the research. This
procedure serves the purpose of disentangling the independent variable in which you may be
interested from other variables with which it is commonly associated so that any effects you
find can more justifiably be associated with the independent variable. Using homogeneous
groupsrestricts the generalizability of the findings only to the specific homogeneous group,
thus reducing external validity of the study.

c. Building Extraneous Variables Into The Design

It may be possible to build relevant extraneous independent variables into the ex post facto
design and investigate their effect through the use of two-way and higher order analyses of
variance.

d. Analysis Of Covariance

Analysis of covariance(ANCOVA) is sometimes used to partially adjust for preexisting

differences between groups in an ex post facto design. Specifically, it adjusts scores on the
dependent variable for any initial differences on the extraneous variable. However, because
the adjustment is only partial, ANCOVA does not “solve” the problem of initial differences
between groups but only reduces it. When interpreting ex post facto research, it is
inappropriate to assume ANCOVA has satisfactorily adjusted for initial differences.
 e. The Role Of Ex Post Facto Research

Given the hazards involved in ex post facto research, many educational researchers believe
they should not engage in this type of research at all. Basically, they contend that it is better
to admit ignorance than to risk reaching incorrect conclusions. Therefore, if they want to
learn anything about relationships between such attribute variables and other variables, the ex
post facto method is their only recourse. An ex post facto study is better than no study at all.
If researchers use appropriate methods of partial control and consider alternative hypotheses,
perhaps they can be correct more often than wrong. Ex post facto research in education has
permitted investigations of the effects of variables such as home background, father absence,
early experiences, disabilities, teacher competence, and others that are beyond the control of
educators. In some instances, ex post facto research has discovered relationships or raised
questions that can later be investigated more systematically in well-controlled experimental
studies. Appropriately used and cautiously interpreted, ex post facto research will continue to
provide a valuable methodology for the acquisition of knowledge. 12

12
Ary,D.,Jacobs,L.C., Sorensen,C.,and Ravazier,A. Introduction to Research in Education.(Boston: Wadsworth Cengange
Learning Products.2010)page 342
 Conclusion

Ex post facto research is used to investigate relationships when the researcher cannot
randomly assign subjects to different conditions or directly manipulate the independent
variable. Ex post facto research begins with subjects who differ on an observed dependent
variable and tries to determine the antecedents (cause) of the difference. Or the researcher
begins with subjects who differ on an independent variable and tries to determine the
consequences of the difference. Although there are many disadvantages of ex post facto
design, it nevertheless is frequently the only method by which educational researchers can
obtain necessary information about characteristics of defined groups of students or
information needed for the intelligent formulation of programs in the school. It permits
researchers to investigate situations in which controlled variation is impossible to introduce.
Attributes such as academic aptitude, creativity, self-esteem, socioeconomic status, and
teacher personality cannot be manipulated and hence must be investigated through ex post
facto research rather than through the more rigorous experimental approach. The possibility
of spurious relationships is always present in ex post facto research. Considering the
possibilities of common cause, reve rsed causality, and possible alternate independent
variables can help educators evaluate such research more realistically. Several partial control
strategies can help researchers avoid gross errors in ex post facto designs, but none can
entirely solve the problems inherent in those designs. Always exercise caution when
interpreting ex post facto results.
 Reference

Ary,D.,Jacobs,L.C., Sorensen,C.,and Ravazier,A. Introduction to Research in Education.

(Boston: Wadsworth Cengange Learning Products.2010)page 332
Ary, Donal. (1982). Pengantar Penelitian Dalam Kependidikan. Surabaya : Usaha Nasional.
Darmadi, Hamid. 2011. Metode Penelitian Pendidikan. Bandung: Alfabeta

Sukardi, Metodologi Penelitian Pendidikan: Kompetensi dan Praktiknya,page 166.

Suharsimi Arikunto, Prosedur Penelitian: Suatu Pendekatan Praktik, (Jakarta: PT. Rineka

Cipta, 2013), page 313.

Richards, Jack, John Platt, and Heidi Weber, Longman Dictionary Of Applied Linguistics,
(Harlow, Essex, England: Longman, . 1985), h. 66

Wicaksono, Andri. 2014. Metode Penelitian Kausal Komparatif Ex Post Facto.

Damayanti, Deni. 2013. Panduan Lengkap Menyususn Proposal, Skripsi, Tesis, Disertasi

untuk Semua Program Studi. Yogyakarta: Penerbit Araska.

Gay,I.R. Geoffrey ,E. Mills, Peter,W. educational research-competence for analysis and
applications(New Jersey: Pearson Education.2012)page 19-20

Handayani,M.et all. Konsep Teori dan contoh desain penelitian Ex Post Facto.
(Jakarta:Universitas Indraprasta Pgri.2015) page 6-7