Sunteți pe pagina 1din 8

Fourth Group Research :

1. Aneka Wahyu Nur sektiani


2. Bambang Iswan

A BRIEF SUMMARY OF
CHAPTER 12
EX POST FACTO RESEARCH
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 sometimes called
causal comparative because its porpuse is to investigate cause nad effect
relationships between independent and dependent variables. Researchers use it in
situations that do not permit the randomizations and manipulation of variables
characteristic of experimental research. Thus, much of the basic rationale for
experimental and ex post facto is the same. They both investigate relationships
among variables and the test hypothesis.

Let us illustrate the differece between ex post facto and the experiment approch by
examining these two approaches tothe same research questions. Consider the
question of the Effect os students anxiety in an achivment testing situation on their
examination performence. The ex post facto approach would involve measuring
that already existing anxiety level at the time of the examination and then
compairing the performance of high anxious and low anxious students. The
weakness of such an approach is that you could necessarily conclude that the
students anxiety prodeced the observed different in achievement examination
performance. Both sets of scores may have been influenced by a third factor, such
as knowledge of the subject matterbeing examined or general self-confidence.
Knowledge or aptitude may be the major cause of both the level of anxiety and the
achievement test result.
In experimental approach to the same problem, the investigator could radomly
assign subject to two exam conditions that are identical in every respect exacp that
one is anxiety arosing and the other is neutral. The experimenter can induce anxiety
by telling the subjects that their final grade may depent on their performance, that
they may find the test to be extremely difficult, or that the test can be used to identify
the incompetent. The neutral group would marely be told that their cooperation is
needed for the experiment. The investigator could randomly assign subjects to the
two conditions. Then, if the anxious group performed better than the neutral group,
it could be concluded that the induced anciety had a facilitating effect on test
performace. If the neutral group performed better than the anxious group, it could
be concluded that the induced anxiety had a debilitating effect. A conclusion could
be legitimately drawn because of the control provided by the random assigment of
groups to treatments and by the experimenters direct manipulation of the
independent variable. Anxiety is one of the few variables that can be either an active
or an attribute independent variable. You can manipulate it actively, as describe
(experimental research), or you classify subjects on the basis of their scores on an
anxiety measure (ex post facto research).

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.
2. Next, select two or more groups to be compared.
3. Determine whether your question requires a proactive or a retroactive
sesign. There are two types of ex post facto research designs proactive and
retroactive.
a. Proactive ex post facto research design begins with subjects grouped on
the basis of independent variable such as father present/father not
present or retained/promoted.
b. Retroactive ex post facto research seeks possible accident causes
(independent variables) for a preexisting dependent variable.
Alternative Explanations in Ex Post Facto Research
1. Common case. A possible of common cause is student employment.
Example having a car may require students to work in order to afford
the car.
2. Reverse causality. Is it possible that poor grades are a cause of car use?
3. Other possible independent variables.

PARTIAL CONTROL IN EX POST FACTO RESEARCH


These strategies provide partial control of the internal valudity problems of
common cause and other posible independent variabels. Among the strategies are
matching, homogeneous groups, building extraneous variables into the design,
analysis of covariance and partial correlation.
a. Matching, a common method of providing partial control in ex post facto
investigation is a match the subjects in the experimental and control groups
on as many extraneous variables as possible. This matching is usually done
a subject to subject basis to form mathed pairs.
b. Homogeneous Groups, you may recall from the disscusion of control in
experimentation that it is possible to control for the effect of a variable by
selecting samples that are ashomogenous as possible on the variable.
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 ther effect through the use of two way and higher order analyses
of cvariace.
d. Analysis of covariance (ANCOVA) is sometimes used to partially adjust for
preexisting differences between groups in an ex post facto design.
Specifically in 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. When interprenting ex post facto research,
it is inappropriate to assume ANCOVA has satisfactorily adjusted for initial
differences.
THE ROLE OF EX POST FACTO RESEARCH
If they want to learn anithing about relationships between such attribute variables
and other variablesin 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.

CHAPTER 13 CORRELATIONAL RESEARCH


A. DEFINITION
Correlational research is non-experimental research that studies the
direction and strength relationships among variables. It gathers data on two or
more quantitative variables from the same group of subjects (or from two
logically related groups) and then determines the correlation among the
variables. Correlational procedures are widely used in educational and
psychological research. They enable researchers to better understand certain
phenomena and to make predictions.
B. USES OF CORRELATIONAL RESEARCH
Correlational research is useful in a wide variety of studies. The most
useful applications of correlation are:
1. Assessing Relationships
Correlational research methods are used to assess relationships and
patterns of relationship among variables in a single group of subjects. For
instance, correlational research is used to answer questions such as:
Is there a relationship between students vocabulary mastery and reading
skill?
2. Assessing Consistency
Correlation can be used to measure consistency (or lack thereof) in
a wide variety of cases. For example:
How consistent are the independently assigned reading score given by the
lecturer and the lecturers assistant at college?
3. Prediction
If two variables are correlated, then you can use one variable to
predict the other. The higher the correlation, the more accurate the
prediction. Prediction studies are frequently used in education. For example:
Is students UN score at SMA related to students GPA (IPK) at College?
C. DESIGN OF CORRELATIONAL RESEARCH
Correlational designs are often valuable for generating hypotheses that
can be further investigated in experimental or ex post facto research. Here the
most popular design for correlational research:
1. Choosing the Problem/Topic
2. Identifies the Problem/Topic
3. Check Literature
4. Make Instruments
5. Tried Out the Instruments
6. Fixed the Sample
7. Gathering the Data
8. Analyze the Data
9. Interpret the Result
D. CORRELATION COEFFICIENTS
Indexes, which correlational research produces, that shown both the
direction and the strength of relationships among variables, taking into account
the entire range of these variables. The sign (+ or -) of the coefficients indicates
the relationships in value form:
1. +1,00 (indicating a perfect positive relationship).
2. -1,00 (indicating a perfect negative relationship).
A number of different types of correlation coefficients are used with
variables that are measured on different types of scales. Kinds of correlation
coefficients:
1. Pearson Product Moment (r)
Pearson product moment correlation coefficients use when the
variables to be correlated are normally distributed and measured on an
interval or ratio scale.
2. Coefficient of Determination (r2)
Coefficient of determination is useful index for evaluating the
meaning of size of a correlation. It also reminds one that positive and
negative correlation of the same magnitude.
3. Spearman Rho ()
Spearman Rho is an ordinal coefficient of correlation. Used when
the data are ranks.
4. The Phi Coefficient ()
Phi Coefficient used when both variables are genuine dichotomies
scored 1 or 0.
E. CONSIDERATIONS FOR INTERPRETING A CORRELATION
COEFFICIENT
Ways to be evaluated the numerical value in correlational research:
1. The nature of the population and the shape of its distribution
The value of an observed correlation is influenced by the
characteristics of the population in which it is observed.
2. Comparison to other correlations
A useful correlation is one that is higher (in either direction) than
other correlations of the same or similar variables.
3. Practical Utility
Always consider the practical significance of the correlation
coefficient. Although a correlation coefficient may be statistically
significant, it may have little practical utility. With a sample of 1000, a very
small coefficient such as 0,08 would be statistically significant at the 0,01
level.
4. Statistical Significance
In evaluating the size of a correlation, it is important to consider the
size of the sample on which the correlation is based.
5. Determining sample size
The Pearson product moment correlation is a form of effect size.
Therefore, it can be used to determine the needed sample size for a
predetermined level of significance and predetermined tolerable probability
of type I error.

6. Correlation and causation


In evaluating a correlational study, one of the most frequent errors
is to interpret a correlation as indicating a cause-and-effect relationship.
Correlation is a necessary but never a sufficient condition for causation.
7. Partial correlation
Partial correlation is a technique used to determine what correlation
remains between two variables when the effect of another variable is
eliminated.
8. Multiple regression
Multiple regression is used to find the relationship between two or
more independent variables and a dependent variable. It yields a prediction
equation that the researcher can use later to predict the dependent variable
for a new group of subjects, when the researcher has information only on
the independent variables
F. FACTOR ANALYSIS
1. Exploratory Factor Analysis (EFA)
EFA is a family of techniques used to detect patterns in a set of
interval-level variables. The purpose of EFA is to try to reduce the set of
measured variables to a smaller set of underlying factors that account for
the patterns of relationships.
2. Confirmatory Factor Analysis (CFA)
CFA used to examine the relationships between a set of measured
variables and a smaller set of factors that might account for them. CFA
assumes relatively precise advance knowledge and allows a researcher to
specify a priori what these relationships might look like and then to test the
accuracy of these formal hypotheses
G. OTHER COMPLEX CORRELATIONAL PRODUCES
Some more sophisticated correlational proce-dures include partial
correlation, discriminant analysis, factor analysis, canonical correlation, path
analysis, and structural equation modeling. It is important to know the type of
research situation in which each of these techniques would be useful.