In contrast to experimental research, non-experimental research allows the investigation or examination of variables or phenomena that cannot be manipulated by the researcher, implying that these variables must be studied as they exist since it is often difficult to alter or control them.
Non-experimental research also encompasses studies that do not employ random assignment of respondents to a specified treatment condition (Belli, n.d.). It is the purpose of this paper to identify and discuss some of existing non-experimental research methods and the type of data that could be used in each of the discussed methods.
One of the main rationales for using non-experimental research, according to Belli (n.d.), is that many variables of interest in the field of social science cannot be controlled or manipulated by virtue of the fact that they are attribute variables. For instance, it is difficult for a researcher to manipulate the social economic status, gender, learning style, thinking pattern, or any other personal characteristic of a respondent because these attributes exist naturally.
Another major reason for using non-experimental research comes from the fact that, in some instances, it would be unethical for the researcher to randomly assign respondents to diverse treatment conditions (Punch & Punch, 2005). For instance, it would by unethical for the researcher to study the effects of drug abuse by randomly assigning respondents to either a drug abusing or a non drug abusing group for a specified time-frame.
It is imperative to note that “…in non-experimental research, groups based on different traits or on self-selection, such as being or not being a smoker, may differ for any other reasons other than the variable under investigation” (Belli, n.d., p. 60).
This is not the case in experimental research, which makes use of the process of random assignment to ensure that individuals included in a sample bears similar characteristics except for the treatment condition in which they are placed.
Consensus about the existing typologies of non-experimental research methods has not been forthcoming, and researchers have resulted in categorizing non-experimental research either based on the purpose of the study or the time-frame delegated for data collection (Belli, n.d.). This paper discusses three different typologies of non-experimental research methods based on the purpose of the study.
The first type is known as descriptive non-experimental research method, in which the fundamental concern for the researcher lies in describing some phenomena of interest or documenting its characteristics (Belli, n.d.). This typology is mainly used to document the status quo of a phenomenon under investigation or to perform a needs assessment in an area that interests the researcher.
According to Punch &Punch (2005), descriptive research designs can utilize nominal data (numbers represent categories), ordinal data (numbers indicate rank order of observation), interval data (numbers represent equal units and intervals between observations can be compared), and ratio data (numbers represent equal intervals or units between absolute zero).
These data can be harnessed into frequency distributions, percentages, averages, central tendencies or observed variability to describe the phenomena of interest (What is Statistics, n.d.).
The second typology of non-experimental research methods is known as correlational research design, in which the fundamental concern of the researcher is to describe the statistical relationship between two or more variables obtained from each subject or phenomena of interest (Punch & Punch, 2005).
Most attitudinal and motivation research heavily relies on correlational research designs to describe and compare the variables of interest. It is imperative to note that in correlational research, data on at least two variables must be obtained from each respondent or phenomena of interest.
According to Punch & Punch (2005), correlational non-experimental research designs are mostly used in “…situations where variation occurs in the independent variables of interest, but where it is not possible to manipulate or control that variation for research purposes” (p. 75).
These authors further posit that correlational non-experimental research designs mostly utilize continuous data, artificially or truly dichotomous data, ordinal data (rank-ordered), and nominal data.
The final typology that will be discussed in this paper is known as explanatory non-experimental research. As acknowledged by Punch & Punch (2005), the primary focus of explanatory research designs lies in attempting to explain how some variables of interest works or why it operates.
The underlying objective of the researcher is often to examine or test a theory about a variable or phenomenon of interest. In most occasions, researchers engaging this method formulate a set of hypothesis from the used theoretical or conceptual framework, which are then tested in an attempt to validate the theory.
According to Johnson & Christensen (2011), the explanatory non-experimental research design mostly utilizes ordinal (rank-ordered) and interval data to explain how a particular variable operates the way it does.
Reference List
Belli, G. Non experimental quantitative research. Web.
Johnson, B., & Christensen, L. (2011). Educational research: Quantitative, qualitative, and mixed approaches, 4th Ed. Thousand Oaks, CA: Sage Publications Inc.
Punch, K.F., & Punch, K. (2005). Introduction to social research: Qualitative & quantitative approaches, 2nd Ed. Thousand Oaks, CA: Sage Publications Inc.
What is statistics? Web.