Quantitative Research Methods - reliability and validity

Validity - this is the extent to which a procedure measures what is intended to measure, there are two types: internal and external validity

Reliability - this is the extent to which a measurement is reproducible

Validity

Internal Validity

This is the ability of the researcher to state that the relationship he or she predicted between the IV and DV does exsiste and is due to the effects of the IV not to extreneous variables. 

The effects of these uncontrolled variables is known as 'secondary variance'. In other words if study is well designed and well controlled and if alternative explanations can be rules out it is said to be internally valid.

External validity

This is the degree to which results generlise toother subjects and situations. Need to select a repersentive sample for this though.

Internal VS External

• By strenghening internal validity, external validity may be compromised
• More careful control = higher interanl validity
• However the more control within the labatory setting the less resemblance to real life and thefore less generalisable
• Needs to be balanced and depends upon aims of the study (basic vs applied research)

Factors that can affect Validity - not all of these will be relevant to every study!!

• Non random assignment - of participants to conditions (all competent participants may end up in one condition)
• Maturation - participants may change over the course of the study (boredom, practice)
• Order effects - performance on a first task is affected by performance on a second (motivation, fatigue)
• Regression towards the mean - baseline differences decrease on follow up
• Time between multiple observations
• Participants drop out and low response rates
• Different time and place
• Different treatment: instructions, practice, trials
• Pre-test sensitization/demand characteristics
• Reactive effects: effects of being observed
• Multiple treatment effects: contamination accross conditions

Dealing with secondary variences

• Eliminate it 
• Hold constant
• Record as co-variate and correct for the effect in the statsics

Eliminating:

• This is the best way of dealing with is 
• Eliminate possible variance due to participant  and researcher expectations

Holding constant:

• Use when elimination isn't possible
• Treat all participants the same and test under the same conditions
• Test everyone in the same room therefore things like tempreture will affect all scores equally

…