Module 5: reliability and validity

Measurement instruments

• measurement reliability

• measurement validity

Both → construct validity

Three ways of repeating measures

• Test-restest reliability→ correlection coefficient

• Inter-rater reliability

• Internal consistency -Cronbach’s alpha

Cronbach’s alpha

>0,95 to similair

>0,90 excellent

>0,80 good

>0,70 acceptable

>0,60 questionable

>0,50 poor

<0,50 unacceptable

Types of validity

• internal validty→ confounbding

• ternal validity → generalize

External validity

• other subjects

• other settings

Good sampling

• representive of the widest population of interest

• large enough to impart adequate statistical power

• exclusion criteria should relate to the question of interest