Textbook Notes
(363,693)

Canada
(158,533)

University of Toronto Scarborough
(18,346)

Psychology
(9,578)

PSYB01H3
(585)

David Nussbaum
(77)

Chapter 5

# Chapter 5.docx

Unlock Document

Psychology

PSYB01H3

David Nussbaum

Fall

Description

Chapter 5
Sampling and survey research
Selecting research participants
1. Studying entire population
a. Census
2. Studying sample of the population
a. Sample
Sample planning
1. Answer two questions
a. What population will you select cases?
b. What method will you use to select cases from this population?
Define the population
2. Cannot generalize findings from one population to another population
a. E.g. from China to Canada
3. Therefore must define population
Define sample components
1. Population
a. The entire set of individuals or other entities to which study findings are to be
generalized
2. Elements
a. The individual members of the population whose characteristics are to be
measured
3. Sampling frame
a. A list of all elements in a population
4. Representative sample
a. A sample that looks like the population from which it was selected in all respects
potentially relevant to the study b. In an unrepresentative sample, some characteristics are over or
underrepresented
c. Random selection of elements maximizes sample representativeness
5. Sample generalizability depends on sampling error
a. The difference between the characteristics of a sample and characteristics of the
population
b. The larger the sampling error, the less representative the sample
c. Thus the less generalizable the findings obtained form that sample
Estimating sampling error
1. Tool for calculating sampling error is called inferential statistics
a. How likely it is that a statistical result based on data from a random sample is
representative of the population from which the sample is assumed to have been
selected
2. Sampling distributions for many statistics have a normal distribution
3. Shape is produced by random sampling error because of chance
a. Differences between the population and the sample that are only due to chance
factors and not to systematic sampling error
b. May or may not result in an unrepresentative sample
c. The magnitude of sampling error due to chance factors can be estimated
statistically
4. Any sample distribution is only one of many possible sample distributions of the
populations
5. Sample statistic
a. The value of a statistic computed from sample data
6. Population parameter
a. The value of a statistic computer using the data for the entire population
b. The sample statistic is an estimate of a population parameter
7. The more cases in the randomly selected sample, the more confident we can be in a
given sample estimate
a. i.e. the confidence limits will be smaller (the extremes of the bell curve) Sampling methods
1. Sampling methods that allow us to know in advance how likely it is that any element of a
population will be selected for the sample are called probability sampling methods
a. You can only make a statistical estimate of sampling error for a probability based
sample
2. Non-probability sampling methods do not allow us know in advance the probability of an
element being selected
3. Probability of selection
a. The likelihood that an element will be selected from the population for inclusion in
the sample
b. As the size of the sample decreases as a proportion of the population, so does
the probability of selection
Probability sampling methods
1. A random sample of 1000 from a population of 1 million is better than a random sample
of 100 from a population of 10,000
2. The four most common methods for drawing random samples are
a. Simple random sampling
b. Systematic random sampling
c. Stratified random sampling
d. Cluster sampling
Simple random sampling
1. Generates numbers based on chance
2. The probability of selection in a true simple random sample is equal for each element
Systematic random sampling
1. Variant of simple random sampling
2. First element is randomly selected from a list, then every nth element is selected
3. Convenient for drawing random samples when population elements are arranged
sequentially
4. Disadvantage a. Possibility of periodicity
i. Sequence varies in some regular pattern that occurs periodically
1. E.g. house selected as every nth element is always the northwest
corner
Stratified random sampling
1. Uses information known about the total population prior to sampling to make the
sampling process more efficient
2. All elements in the population are distinguished according to their value on some
relevant characteristic
a. This characteristic forms the sampling strata
i. Elements are sample randomly from within each strata
3. Example:
a. Race are the population of interest
i. Each race forms a strata
1. From there randomly sample from each strata
4. This method is more efficient than drawing a simple random sample because it ensures
appropriate representation of elements across strata
5. Two types of stratified sampling
a. Proportionate
i. Sample is selected so that the distribution of characteristics in the sample
matches the distribution of corresponding characteristics in the population
b. Disproportionate
i. Sample does not match the distribution of the population
Cluster sampling
1. Useful when a sampling frame of elements is not available
2. Applicable often for large populations spread across a wide geographic area
3. Cluster is a naturally occurring mixed aggregate of elements of the population
a. With each element appearing in one cluster i. Example:
1. Country is large cluster
a. States are smaller clusters (select for random states)
i. Cities are smaller clusters within states (select for
random cities)
1. Schools are smaller clusters within cities
(select for random schools)
a. Elements randomly sampled from
within schools
Nonprobability sampling methods
1. Availability sampling
a. Does not use random selection procedure
b. Do not know if a sample is representative of the larger po

More
Less
Related notes for PSYB01H3