Lakhasly

NONEXPERIMENTAL RESEARCH
In nonexperimental quantitative research, the researcher identifi es variables
and may look for relationships among them but does not manipulate the variables. Major forms of nonexperimental research are relationship studies including ex post facto and correlational research and survey research.
Ex post facto research is similar to an experiment, except the researcher
does not manipulate the independent variable, which has already occurred in
the natural course of events. The researcher simply compares groups differing
on the preexisting independent variable to determine any relationship to the dependent variable. Because there is no manipulation or control of the independent variable, one must be very careful regarding the conclusions that are drawn
about any observed relationship. The variables in this type of research differ in
kind rather than in amount. Later, we refer to these as categorical variables. For
example, to answer the question, “What is the effect of part-time work on school
achievement of high school students?” one would use the ex post facto method.
The researcher would not manipulate the lives of high school students by having
some take part-time jobs and others not; instead, the researcher would identify
students who already work part-time and compare their achievement with that
of students who do not work. Because researchers lack control over many factors, they must be especially careful in interpreting the results of ex post facto
investigations. They do not establish a simple causal relationship among the variables of a study. We discuss ex post facto research in detail in Chapter 12.
Correlational research gathers data from individuals on two or more variables
and then seeks to determine if the variables are related (correlated). Correlation
means the extent to which the two variables vary directly (positive correlation)
or inversely (negative correlation). The degree of relationship is expressed as a
numeric index called the coeffi cient of correlation. Correlational research might
ask about the relationship between the quality of writing samples produced by
incoming college freshmen and their academic performance during the freshman
year. Also, one might investigate the relationship between performance on a language aptitude test and success in a high school foreign language course.
Both ex post facto and correlational research investigate relationships between
variables. The major distinction between the two is that in ex post facto research
one categorizes the participants into at least two groups on one variable and then
compares them on the other variable. In correlational research, a researcher
deals with one group of individuals measured on at least two continuous variables. We discuss correlational research in Chapter 13.