Overview
In this module, we will discuss how researchers use sample data to draw conclusions about larger populations. We focus on the use of sampling in quantitative research, and consider three factors that come together to determine the representativeness of a sample.
As always, to help you better understand and engage with the ideas that we are covering in the modules ahead, I strongly encourage you to discuss the course material in the class discussion boards.
When you have finished this module, you should be able to do the following:
- Distinguish between a population and a sample, and between census data and sample data;
- Explain the logic of drawing samples from larger populations;
- Explain what a sampling frame is, its role in sample representativeness, and the challenges in ensuring an accurate sampling frame;
- Distinguish between probability and non-probability sampling;
- Describe common probability sampling techniques;
- Define confidence intervals;
- Explain why sample size matters for probability samples; and
- Critically assess sampling approaches when reading social science research studies.
- Read Chapter 8 (pages 152-168) of the 3rd edition of our textbook or Chapter 5 (pages 83-101) of the 4th edition of our textbook. Create self-study flashcards for the chapter.
- Watch mini-lectures “Introduction to Sampling”, “Sampling Frames”, “Sample Selection”, “Sample Size”, and “Critically Assessing Samples.”
- Complete Learning Activity.
- Population
- Census
- Sampling
- Population parameter
- Statistic
- Sampling error
- Representative sample
- External validity
- Sampling frame
- Probability sampling
- Nonprobability sampling
- Simple random sampling
- Stratified sampling
- Cluster sampling
- Confidence interval
- Margin of error
- Homogeneity
- Heterogeneity
- Chapter 8 (pages 152-168) in Berdahl, Loleen and Keith Archer. Explorations: Conducting Empirical Research in Canadian Political Science (Third Edition). Oxford University Press OR Chapter 5 (pages 83-101) of Berdahl, Loleen and Jason Roy. Explorations: Conducting Empirical Research in Canadian Political Science (Fourth Edition). Oxford University Press.
Learning Material
Learning Activity
Random Sampling Research Design
- Design a survey sampling plan for a study of current university students at the University of Saskatchewan. Assume that there are 10,000 students and that you have a list with all of their names and email addresses. Using an online sample size calculator (e.g., https://www.surveymonkey.com/mp/sample-size-calculator/), identify how many students you need to include in your sample to have a +/-3% margin of error at a 95% confidence level. In 300-500 words, explain how you could randomly sample this population to get the desired number of responses, and discuss potential sources of bias that could occur in your study. In your response, use at least two terms covered in the module (readings and/or videos) and be sure that all terminology is used correctly. Proofread carefully.
- Post your sampling plan in your Learning Activity Discussion Board.
- Provide a constructive response to at least one of your fellow group members’ posts. A constructive response is one that (a) uses supportive language to (b) identify for the author an area in which the work can be strengthened. For example, it may identify an issue where the wording is unclear or a point where terminology is used incorrectly, or suggest ideas for examples or ways to strengthen the argument, or let the author know of questions that the work raised for them. A constructive response goes beyond ‘I agree’ or ‘that is interesting’ to assist the author in improving the work. It should provide feedback that is intended to assist the author of the learning activity in improving their work
Reminder: At the end of Module 4, you are required to select one learning activity for submission from Modules 1-4. You can use the feedback that you receive in the group forum to revise your selected learning activity prior to submission.
Glossary
census an enumeration or a record of the full population.
cluster sampling a probability sampling technique in which the researcher divides the population into a number of subgroups (i.e. clusters) and then randomly selects clusters within which to randomly sample.
confidence interval the estimated range of values within which the population parameter is likely to fall.
external validity the extent to which the findings drawn from the cases under examination may be used to make generalizations about phenomena outside the original study.
heterogeneity the degree of difference between a population and the variable of interest.
homogeneity the degree of similarity between a population and the variable of interest.
margin of error a range around the estimate that likely contains the population parameter; used by researchers to state their sample statistics as a confidence interval
population in research, the group that a researcher wishes to generalize about
population parameter population characteristics, expressed in numeric terms when the responses of each member (or case) of the population are measured.
random selection a selection technique in which all cases in a population have an equal opportunity for inclusion in the sample.
representative sample a sample that accurately reflects the larger population from which it was drawn.
sample a record of a subset of a population.
sampling choosing a number of cases or available texts from a larger population rather than analyzing the entire population.
sampling distribution the theoretical distribution of a sample statistic (e.g. the mean) for a given sample size.
sampling error the difference between the sample statistic and the population parameter.
sampling frame a list of all the units in the target population.
simple random sampling the process by which every case in the population is listed and the sample is selected randomly from this list.
statistic a numeric estimate of the population parameter.
stratified sampling a probability sampling technique in which the researcher breaks the population into mutually exclusive subgroups, or strata, and then randomly samples from each group.
Note: Unless otherwise stated, glossary source is Berdahl, Loleen and Keith Archer. 2015. Explorations: Conducting Empirical Research in Canadian Political Science (Third Edition). Oxford University Press.
References
Berdahl, Loleen and Keith Archer. 2015. Explorations: Conducting Empirical Research in Canadian Political Science (Third Edition). Oxford University Press.
Grant, Tavia. 2013. “Canadian income data 'is garbage' without census, experts say.” The Globe and Mail. https://www.theglobeandmail.com/report-on-business/economy/without-census-data-on-canadian-income-garbage-experts/article14701515/
Lazerwitz, Bernard. 1968. “Sampling Theory and Procedures.” In Methodology in Social Research, edited by Hubert M. Blalock, Jr., and Ann B. Blalock, 278–328. New York: McGraw-Hill.
Potter, Andrew. 2010. Sometimes a gaffe is more than a gaffe. MacLeans. https://www.macleans.ca/general/sometimes-a-gaffe-is-more-than-a-gaffe/
Statistics Canada. 2013. Statistics: Power From Data! Ottawa: Statistics Canada. http://www.statcan.gc.ca/edu/power-pouvoir/toc-tdm/5214718-eng.htm (accessed 7 October 2013).