Internal validity is a scientific concept that addresses the relationship between two variables. It refers to the extent that a study can rule out or make unlikely alternate explanations of the results.
Definition of Internal Validity
The purpose of conducting research is to arrive at valid and reliable conclusions about a variety of topics.
We look at both the interactions between variables and how one variable impacts another variable. The dependent variable is the item that we plan to measure and change, while the independent variable is the intervention that we manipulate and that causes the change. For example, if we wanted to see if a certain medication promoted weight loss, the drug would be the independent variable, and the dependent variable would be the amount of weight loss achieved.Scientific research cannot conclude with absolute certainty that the independent variable caused the change in the dependent variable.
Internal validity is a scientific concept that addresses the relationship between two variables. It refers to the extent that a study can rule out or make unlikely alternate explanations of the results. Influences other than the independent variable that might explain the results of a study are called threats to internal validity.
Threats to Internal Validity
There are several threats to internal validity that may exist in an experiment.
History is a threat to internal validity; it refers to any event other than the independent variable that occurred in or out of the experiment that may account for the results of the experiment.
It refers to the effects of events common to subjects in their everyday lives. Events that occurred in the weather, in the news or in the subject’s personal lives could alter their performance in an experiment. In our weight loss example, the subject’s eating habits and activity level outside of the experiment could impact the amount of weight loss or gain.
Changes over time could also result from processes within the subjects themselves. Maturation refers to those processes such as growing older, growing stronger and even growing tired and impatient.
Maturation will only be a problem if the design of the experiment does not identify and separate these effects. It is very common to see both history and maturation together, although it is not required.
Attrition refers to the loss of subjects over time.
Some studies last days, weeks and even months, and not all subjects remain for the entire duration. When a subject quits, moves, drops out or is removed from the research, it is called attrition.
Testing refers to what happens to our test performance when we practice or take a test again.
Familiarity with the test could influence the performance on the second testing. Changes in the final scores may be a result of repeated testing.
Instrumentation simply refers to the actual changes in the measuring procedures or the measuring device, rather than any changes in the person over time. This is one of the key reasons that we use standardized and formal testing procedures when conducting any type of assessment.
Statistical regression refers to the tendency for extreme scores to revert or regress toward the mean (or average) of a distribution when a measure is readministered. For example, if an event occurs during the day just before or just after the hour, one could forecast that the event will next occur on the hour in the future.
Selection biases refer to the differences between groups that are identified before any experiment begins. This is one of the most obvious threats to internal validity because any time you are looking at preexisting groups or intact groups, they may differ from one another in a number of ways. To deal with this threat, we use randomization and deliberately assign or exclude subjects based on specific variables, such as age, gender, ethnicity, etc.
Diffusion of Treatment
Diffusion of treatment occurs when the intervention given to one group may accidentally or inadvertently be given to all or some subjects in another group, usually the control group that was not supposed to receive any intervention at all.
Using our weight loss example again, if the control group is watching the news on television and overhears that increasing caffeine intake could assist with weight loss, the control group may engage in that behavior as well, and similar results may be achieved.
In an experiment, the dependent variable is the item that we plan to measure and change, while the independent variable is the intervention that we manipulate and that causes the change. Internal validity is a scientific concept that addresses the relationship between two variables. It refers to the extent that a study can rule out or make unlikely alternate explanations of the results.Influences other than the independent variable that might explain the results of a study are called threats to internal validity. Threats to internal validity include history, maturation, attrition, testing, instrumentation, statistical regression, selection bias and diffusion of treatment.
Once this lesson ends, you could seize the opportunity to pursue these objectives:
- Differentiate between the dependent and independent variables in a study
- Define internal validity
- List eight threats to internal validity