Research Methods Week 4

Research Methods Week 4

1. Causation - sufficiency and necessity

   • Need to satisfy necessary and sufficient criteria in order to make true claim about causality
   • Sufficient - Y is adequate to cause X
     • Necessary - Y must be present to cause X
   • To be good at psychology you need to be good at RM - RM is necessary to be good at psychology. But you must be good at other things also, so RM is not sufficient to be good at psychology
   • Completing and passing a UG degree at UoM will obtain you a BSc - the degree is sufficient in order to gain a BSc - but you can gain it at other unis on other courses - doing psychology at UoM not necessary to obtain BSc
     • To obtain full marks on the exam, it is necessary to answer every question correctly - it is also sufficient to answer every question correctly

2. Precision & Accuracy

   • Precision - exactness (consistency)
   • Accuracy - correctness (truthfulness)

3. Reliability & Validity

   • Reliability - precision (consistency) - extent to which our measure would provide the same results under the same conditions
     • Forms of reliability
       • Test-retest - measures fluctuations from one time to another - important for constructs we expect to be stable - e.g personality type
       • Inter-rater - measures fluctuations between observers
       • Parallel - if we administer different versions of our measure to same Ps, would be obtain same results?
         • Beware of order effects
       • Internal consistency - determines whether items are measuring the same construct - split-half reliability - questionnaire items split into 2 groups and the halves are administered to Ps on separate occasions
   • Validity - accuracy (truthfulness) the extent to which we are measuring the construct we're interested in

4. Forms of Validity

   • Content validity - does our test measure the construct fully?
   • Face Validity - does it look like a good test?
   • Criterion validity - does the measure give results which are in agreement with other measures of the same thing - e.g does the RM exam relate to lab grades?
   • Concurrent - comparison of new test with established test
   • Predictive - does the test predict outcome on another variable?
   • Construct validity - is the construct we are trying to measure valid? i.e does the construct itself exist? Supported by cumulative research collected over time
     • In the mean time construct validity can be assessed in terms of convergent validity - correlates with tests of same/related constructs, and discriminant validity, doesn't correlate with tests of different or unrelated constructs

5. Samples & Populations

   • Populations - entered collection of people we are interested in - sharing a common characteristic, defined by population parameters, vary in size
   • Sample - a selection of individuals from the larger population - defined by sample statistics, sample stats are used to infer pop. parameters
     • Why sample? Time, money, access, sufficiency
     • Random - gold standard, each member of population has chance of being selected
     • Systematic - draw from pop. at fixed intervals - problematic in populations with a period function
     • Stratified - proportional - specified groups appear in numbers proportional to size in pop. or disproportional, specified groups which are not equally represented in pop. selected in equal proportions
     • Cluster - researcher samples an entire group or cluster from a population of interest
     • Opportunity/convenience - can lead to biased sample
     • Snowball sampling - recruit small numbers of Ps and then get them to recruit others - biased sample but useful if you want to recruit specific populations

6. External Validity

   • Refers to ability to generalise results
     • Population validity - is it representative?
     • Ecological - does it reflect naturally occurring behaviour?