Class Notes (837,447)
COMM 2002 (65)
Lecture

# COMM December 2, 2013.docx Premium

4 Pages
42 Views

School
Department
Communication Studies
Course
COMM 2002
Professor
Heather Pyman
Semester
Fall

Description
Sampling •Representative Samples o Populations - Parameters o Samples - Statistics o Probability sampling vs Non-probability sampling • How do we know that samples we select are representative of the population? • Collect data in a random sampling method -> key is that each person has an equal and likely chance to be selected for the sample  Have to know the population and then have to be able to illustrate and prove that every person ahd the same chance in being selected for tehs ample as everyone else does • When we collect them, the analysis ran creates statistics: descriptive statistical analysis • Inferential statics: Generalizing from a sample to a situation  If we are going to say the sample is representative of the population, we have to conduct a probability based method  Random, representative or probability based method -> know the probability of any one person being chosen form that sample  Researchers are not interested in the population • Quality research -> ontological and epistemological is more attitude and behaviour rather than inferring form the research that they've done form a population -> non probability, sometimes do quantitative work •Quality of a probability sample o Representative - allows for generalization from sample to population o Dependent on three things • Accuracy f the sampling frame • Sample size • Sampling method  Willing to take a 5% risk to what the actual sample is  How to collect the sample; accurately describe the population -> determine a sampling frame, frame where we are going to collect the sample • If the population is all students from Carleton, could be the registar's list, student IDs, etc. •Probability sampling o Inferential statistical tests o Sample statistics can be used to estimate population parameters o Standard error (SE): Estimate f discrepancy between sample mean and population • Standard deviation of sample distribution to estimate the discrepancy between the population value and other o Remember that sample statistics are not likely to be actual population values •Unrepresentative Probability Samples o Not using a random sampling method o Sampling frame is too small o Non-response o Sampling error • Frame could be too small -> not enough sample to represent all of the population  Could happen when elements of the population [in the sampling frame] are not known to the researcher  Researchers are looking for a 50% response rate, that is the people who did not answer, are not significant and are not a larger amount of group that did answer the survey • Sample size o Absolute size matters more than relative size o The larger the sample, the more precise and representative it is likely to be o As sample size increases, sampling error decreases, BUT there is a point of diminishing returns • If you decrease a number, would be looking at a sampling error, cut the sample size into a quarter means you've reduced error by 50% • If you went to a sample size of 100 from 1000 then it is error rates of 3% • Think of sample as absolute -> no • Things to be determined when determining sample size o Homogeneity of the sample o Number of variables in the study o The desired degree of accuracy o Time and cost o Non-response o Kind of analysis to be carried out • What we need to consider during the data analysis is how much we want to break the survey down •
More Less

Related notes for COMM 2002
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.