Internal Validity

History

A threat to internal validity due to life events in the subjects life that may be partially or fully responsible for changes recorded in the del variable after the in depend variable is introduced. History is the events that occur in addition to the experimental variable between the first and second measurements. Example: treating a severe phonological disorder in a pre k student. But she gets tubes during the 3rd week of treatment. The progress made can't alone be attributed to the treatment itself.

Instrumentation

A threat to internal validity due to mechanical instruments (audiometer) or paper and pencil scoring. Examples: if the audiometer is recalibrated in the middle of the study, then the pretest and post test hearing were screened on different calibrations and may not be as consistent. Or,if the same judges glottal fry pre and post treatment, they may be more susceptible to hearing glottal fry by the post testing, making the pre and post screens less consistent.

Attrition (mortality)

A threat to internal validity in which the experiment loses participants as the study progresses. Has an effect on final results. Big issue for group designs. If a lot of people drop out of the severe group, and less drop out of the NML group, when you compare the two, the data may be non representative. This doesn't affect single subject designs.

Statistical Regression

A threat to internal validity in which the progress may not be fully due to the treatment itself but rather natural consequences. refers to the behavior that goes from an extreme high or low point, to an average level. Example: A girl is very hoarse after cheerleading at a football game. She is studied and begins tx while she is at her worst...Therefore the some of the seen progress could be due to natural recovery as the VFs heal, not just the intervention.

Testing

A threat to internal validity referring to a change that occurs in a dependent variable simply because it has been measured more than once. Taking 5 practice tests before the real test may be the reason for a final improved score.

Interaction of Factors

A threat to internal validity referring to any combination of the above factors.

Subject Selection Biases

A threat to internal validity referring to the subjective selection of who participates in the study. Say you have a group design where in the end, there were differences between the two groups and the cause seems to be the intervention. However, if the two groups were different to begin with then the final results showing a difference may not actually be because of the independent variable.

Maturation

A threat to internal validity referring to uncontrollable biological or other changes within the participant affecting the dependent variable. If you begin treating kindergarten kids with lang stimulation in Fall of 2016, then test in may of 2017, the progress may not just be due to the lang stimulation program, but to their natural development.

Internal Validity

degree to which data in a study reflect a true cause-effect relationship. A FAILURE TO ESTABLISH A TRUE CAUSE-EFFECT RELATIONSHIP. If there is strong internal validity, then the dependent variable was effected only by the manipulation of the independent variable. No influence of confounding variables on the dependent variable.

Threats to internal validity

instrumentation, history, Statistical regression, maturation, attrition, testing, subject selection biases, interaction of factors