Research Methods: Validity

Validity

When researchers make conclusions from data, they have to be sure that they are justified. Researchers have to be sure that the data can be trusted to represent what they intended it to. Measures must be reliable if they are also going to be valid.

Validity

• Validity refers to whether the test/measure is measuring what it claims to measure. 
• Internal validity
• External validity.

Internal Validity

• Ability to study to test the hypothesis that it was designed to test. (does DV measure what is was intended to measure?)

Potential problems and controls for internal validity:

Experimental Design - 

• if a repeated measures design is used there is a danger of order effects (e.g practice effects/boredom). If an independant groups design used then ppt variables (extraneous variables) occur.
• Control - order effects overcome by counterbalancing conditions. Ppt variables overcome by using matched pairs design.

Poor Operationalisation of Variables - 

• many psychological variables are abstract and is possible to choose an inappropriate method of measuring them.
• Control - clearly defining all research variables or use measures and scales that have been validated in previous studies.

Extraneous Variables - 

• impossible to know if the IV or extranous variables caused changes in DV
• Control - studies need to be carefully controlled. All variables except IV kepts constant/controlled. Standardised procedures help with control. (makes sure each ppt tested in same way)

Demand Characteristics - 

• …