If so, then internal validity might not be as strong.

Again, this criticism does not question the internal validity of the original studies.

If the study has internal validity then the findings are held true within the sample.

This makes it very difficult to appreciate the internal validity of the models.

Using unequal groups can also be a threat to internal validity.

In other words, descriptive research can be said to have a low requirement for internal validity.

Both internal and external validity are not captured in a single experiment.

Thus, if executed properly, controlled studies have high levels of internal validity.

Statistical regression, history and the participants are all possible threats to internal validity.

Test items are then selected or eliminated based upon which will result in the strongest internal validity for the scale.