VB SC 050S Lecture Notes - Lecture 3: Sample Size Determination, Simple Random Sample, Nonprobability Sampling

• School: PSU
• Department: Veterinary and Biomedical Sciences
• Course: VB SC 050S
• Professor: Griel Lester

Document Summary

Sampling: de ne the sampling unit, probability sampling method. Every individual in the population has an equal chance of being included in the study. Strati ed random sample: putting a random sample on top of a non probability sample. Cluster sample; dividing it up (all dogs in randomly selected neighborhoods: non probability sampling method. Do not rely on formal random techniques to identify units to be included. Judgement (we shouldn"t include newborn pups), convenience (can"t nd all the dogs in one area so focus on dogs going to vets), purposive sampling (reason, like only take dogs with a certain medical condition: sample size determination. N = 4 x s2 / l2. Accuracy of diagnostic testing procedures: sensitivity, speci city, predictive values. Measures of correct classi cation of individuals into two categories compare to a. Sensitivity: a / (a+c) = 850/1000 = 85%; speci city: d / (d+b) = 1350/1750 = ~75%