This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Lecture 04 (Jan 28, 2009) What to Measure- what are we going to measure? Depends on what we want to do with the measure • Sampling and generalizability Population vs. sample Sampling techniques – procedures for deciding which examples of the population you will measure A sample is a smaller grp of examples chosen from the population that we are actually going to measure; sample and population are often hard to differentiate We can define the nature of the population; therefore the sample becomes blurred This is important b/c what we want to do is draw conclusions from our sample (Generalization: we want to generalize the info from the sample to the population) The sample might not generalize well to the entire population i.e. What some ppl care for older adults vs institution? our sample is the elderly in institution, and find out their characteristics (problem: do elderly ppl in institution represent OAs as a whole? Not really b/c those in homes are often sicker and more medical problems) sample is represented if it is similar in all measured respects of the population. When the sample is represented, it is safe to generalize. Generalization and representative To generalize from sample to population, need to know if sample is representative of the population e.g) Galop accurately predicted the winners of 3 US presidential election prior to 1940. But, next time, he didn’t get his prediction right. Where did he screw up? One problem was that he didn’t use a sample that was represented of the population. -how is it that we select our samples? This is sampling techniques, the ways we choose to choose sample. • Types of samples Non-probability sampling (Every member of population does not have equal opportunity of being selected for the sample, and therefore not represented of the population, there are issues of representativeness. People use this sample due to convenience. Haphazard or convenience sampling simply observing the members of the population that are there, convenient. Self-selected sample: This convenient sample is chosen by the people in the study, rather than the experimenter, ppl choose to be in the sample. Quota sampling we know something abt the population (X females, Y males), then try to match the percentages of the sample (51 females, 49 males). You choose these ppl in a haphazard way, out of convenience. (This is where Galop went wrong, he used the wrong census data to make his prediction) Probability sampling (Every member of the population has a equal chance of being...
View Full Document
- Spring '09