Non-experimental research is one that lacks the manipulation of an independent variable, the random assignment of participants to conditions or orders of conditions, or both.
In a sense, it is unfair to collectively define this broad and diverse set of approaches for what they are not. But doing so reflects the fact that most researchers in psychology consider the distinction between experimental and non-experimental research to be extremely important.
This distinction is because, although experimental research can provide strong evidence that changes in an independent variable cause differences in a dependent variable, non-experimental research generally cannot. However, as we shall see, this inability does not mean that non-experimental research is less important than experimental research or inferior to it in any general sense.
When to use non-experimental research
Experimental research is appropriate when the researcher has a specific research question or hypothesis about a causal relationship between two variables, and it is possible, feasible, and ethical to manipulate the independent variable and randomly assign participants to conditions or condition orders. It is therefore logical that non-experimental research should be appropriate – even necessary – when these conditions are not met. There are many ways to prefer non-experimental research.
The research question or hypothesis may refer to:
A variable rather than a statistical relationship between two variables (for example, what is the accuracy of people’s first impressions?)
A non-causal statistical relationship between variables (for example, is there a correlation between verbal intelligence and mathematical intelligence?).
Causal relationship, but the independent variable cannot be manipulated or participants cannot be randomly assigned to conditions or condition orders (e.g., does damage damage to a person’s hippocampus the formation of long-term memory traces?)
Broad and exploratory, or can it be about what a particular experience looks like (for example, what is it like to be a working mother who has been diagnosed with depression?)
Choosing between experimental and non-experimental approaches
Again, the choice between experimental and non-experimental approaches is often dictated by the nature of the research question. If it is a causal relationship and involves an independent variable that can be manipulated, the experimental approach is usually preferred. Otherwise, the non-experimental approach is preferred. But the two approaches can also be used to address the same research question in a complementary way.
For example, non-experimental studies establishing that there is a relationship between violent television watching and aggressive behavior have been supplemented by experimental studies confirming that the relationship is causal (Bushman and Huesmann, 2001). Also, after his original study, Milgram conducted experiments to explore the factors that affect obedience. He manipulated several independent variables, such as the distance between the experimenter and the participant, the participant and the Confederate, and the place of study (Milgram, 1974).
Types of non-experimental research
Non-experimental research is divided into three broad categories: single-variable research, correlational and quasi-experimental research, and qualitative research.
Research with a single variable
First, the research may be non-experimental because it focuses on a single variable and not on a statistical relationship between two variables. Although there is no widely shared term for this type of research, we will call it single-variable research. Milgram’s original study of obedience was non-experimental in this regard.
He was primarily interested in one variable — the extent to which the participants obeyed the researcher when he told them to give a discharge to the Confederate — and observed all the participants performing the same task under the same conditions. Loftus and Pickrell’s study described at the beginning of this chapter is also a good example of single-variable research.
The variable was whether participants “remembered” experiencing slightly traumatic childhood events (e.g., getting lost in a shopping mall) that they had not actually experienced but were repeatedly asked by research. In this particular study, nearly a third of participants “remembered” at least one event. (Like Milgram’s original study, this study inspired several later experiments on factors that affect false memories.)
As these examples demonstrate, research with a single variable can answer interesting and important questions. What it cannot do, however, is answer questions about the statistical relationships between variables. This detail is a point that beginner researchers sometimes overlook. Imagine, for example, a group of research method students interested in the relationship between children being bullied and children’s self-esteem.
The first thing these researchers would come up with is to get a sample of high school students who have been bullied and then measure their self-esteem. But this design would be a study of a single variable with self-esteem as the only variable. Although I would tell researchers something about the self-esteem of children who have been bullied, I wouldn’t tell them what they really want to know, which is how the self-esteem of children who have been bullied compares to the self-esteem of children who have not been bullied. Is it lower? Is it the same? Could it be even higher? To answer this question, your sample would also have to include high school students who have not been bullied, thus introducing another variable.
The research may also be non-experimental because it focuses on a statistical relationship between two variables, but does not include the manipulation of an independent variable, the random assignment of participants to conditions or orders of conditions, or both. This type of research takes two basic forms: correlational research and quasi-experimental research.
In correlational research, the researcher measures the two variables of interest with little or no attempt to control the foreign variables and then evaluates the relationship between them. A research method student who finds out if each of the high school students has been the victim of bullying and then measures the self-esteem of each of them is conducting correlational research.
In quasi-experimental research, the researcher manipulates an independent variable but does not randomly assign participants to conditions or orders of conditions. For example, a researcher may launch an anti-bullying program (a type of treatment) at a school and compare the incidence of bullying at that school with the incidence at a similar school that does not have an anti-bullying program.
The last way research can be non-experimental is that it can be qualitative. The types of research we have analysed so far are all quantitative, referring to the fact that the data consist of numbers that are analysed using statistical techniques. In qualitative research, the data are often non-numerical and therefore cannot be analyzed using statistical techniques.
Rosenhan’s study of people’s experience in a psychiatric ward was primarily qualitative. The data were the notes taken by the “pseudopacientes” – the people who pretended to have heard voices – along with their hospital records. Rosenhan’s analysis consists mainly of a written description of the experiences of pseudopacientes, supported by several concrete examples.
To illustrate the tendency of hospital staff to “depersonalize” their patients, he noted, “Upon being admitted, I and other pseudo-patients underwent initial physical examinations in a semi-public ward, where staff members engaged in their own affairs as if we were not there” (Rosenhan, 1973, p. 256). Qualitative data have a separate set of analysis tools depending on the research question. For example, thematic analysis would focus on the issues that arise in the data or conversation analysis would focus on how words were said in an interview or discussion group.
Revised internal validity
Recall that internal validity is the degree to which a study design supports the conclusion that changes in the independent variable caused any observed difference in the dependent variable. Experimental research tends to be the highest because it addresses directionality and third variable problems by manipulating and controlling foreign variables through random assignment. If the mean score of the dependent variable in an experiment differs between conditions, it is very likely that the independent variable is responsible for that difference.
Correlational research is the lowest because it does not address either problem. If the mean score on the dependent variable differs between the levels of the independent variable, it could be that the independent variable is responsible, but there are other interpretations. In some situations, the direction of causality could be reversed. In others, there could be a third variable that causes differences in both the independent and dependent variables. Quasi-experimental research is somewhere in between because manipulation of the independent variable solves some problems, but the lack of random assignment and experimental control does not solve others.
Imagine, for example, that a researcher finds two similar schools. In this regard, it launches an anti-squaraction programme in one of them. It then finds fewer incidents of harassment in that “treatment school” than in the “control school.” There is no directionality issue because clearly the number of bullying incidents did not determine which school received the program. However, the lack of random assignment of children to schools could mean that students in the treatment school differ from students in the control school. This could explain the difference in bullying.
Our specialists wait for you to contact them through the quote form or direct chat. We also have confidential communication channels such as WhatsApp and Messenger. And if you want to be aware of our innovative services and the different advantages of hiring us, follow us on Facebook, Instagram or Twitter.
If this article was to your liking, do not forget to share it on your social networks.
Bushman, B. J., & Huesmann, L. R. (2001). Effects of televised violence on aggression. In D. Singer & J. Singer (Eds.), Handbook of children and the media (pp. 223–254). Thousand Oaks, CA: Sage.
Milgram, S. (1974). Obedience to authority: An experimental view. New York, NY: Harper & Row.
Rosenhan, D. L. (1973). On being sane in insane places. Science, 179, 250–258.
You may also be interested in: Probabilistic and Non-Probabilistic Sampling