PSYC 412 Lecture Notes - Lecture 3: Discriminant Validity, Random Assignment, Conduct Disorder

• School: McGill University
• Department: Psychology
• Course: PSYC 412
• Professor: Melanie Dirks

Document Summary

Inter-rater reliability: two people coding the measure should come up with the same result. Internal consistency: items tapping the same construct are related to each other. Test-retest reliability: measure taken at two different time points show the same results. Validity: if the measure is measuring what we want it to. Sample (group that is being studied) vs. population (group we are interested in) We want the sample to be representative of the population. The best way is random sampling: everyone in the population has an equal chance of being selected. Identify rates of a problem in the population, may be interested in: New cases over a time period (incidence) All cases during a time period (prevalence) Sampling and measurement of disorder are critical for this type of study; epidemiologists need to be able to generalize their findings to the population of interest. Need random sampling and not likely to use a sample of convenience (sample who is readily available)