Internal validity

Two considerations affect an experiment's internal validity:

1 the influence the treatment had on the actual outcome

2 the influence of extraneous factors.

Extraneous factors can take a number of forms (see Figure 9.2).

History and maturation The changes associated with the passage of time are described as history and maturation. The influence of history is typified by changes in current events external to the experiment but that still may have affected its outcomes. The impact of history becomes greater as the length of the experiments increases. Maturation is primarily concerned with changes occurring within the individual subjects.

Repeated testing This applies in experiments where the same people are tested a number of times. As they become familiar with the testing procedure, their performances may change solely because of this familiarity.

Impact of the researchers themselves The researchers themselves may cause some of the change. Their very presence may influence the actions of the subjects being observed.

History and maturation

RfPefated researchers testing themselves

Mortality of participants

Selection Regression errors effects


Extraneous factors influencing validity

Mortality of participants One problem with experiments carried out over a long period is mortality, meaning that, for one reason or another, some of the original participants 'drop out'. The problem is often encountered with consumer panels. There is no way of knowing whether the test units lost would have responded similarly to the treatment as those units that remained.

Selection errors The test units may be chosen in a manner that biases the results. They may not be representative of the desired population or the treatments are assigned to them in such a way that they influence the outcome.

Regression effects Regression effects occur when test units are chosen on the basis of their extreme score or performance on some earlier test. The outliers, especially those with very low performance, tend to move to a more average position over time.

All these factors represent causes for error in the experiments. Many of them can be controlled if the appropriate experimental design is used. Experiments should be designed primarily to control those extraneous factors most likely to seriously influence the outcome and not try to eliminate all sources of error.

Online Survey Champion

Online Survey Champion

There are people all over the world trying to find ways to make money online. From stay at home moms looking to make a few extra dollars to college students and entrepreneurs, the allure of making your own hours and working from home or from the local coffee shop is very appealing.

Get My Free Ebook

Post a comment