Textbook Notes
(362,796)

Canada
(158,054)

University of Toronto Scarborough
(18,305)

Psychology
(9,545)

PSYB01H3
(585)

David Nussbaum
(77)

Chapter 5

# Chapter 5 Notes 97-03.doc

by
OneClass2287

Unlock Document

University of Toronto Scarborough

Psychology

PSYB01H3

David Nussbaum

Summer

Description

PSYB01 Chapter 5 – Sampling and Survey Research
Selecting Research Participants
- Sampling: the selection of individuals or other entities to represent a larger population
- Census: to study the entire population of interest
- Better to survey a limited number so there are more resources for follow-up
procedures
Sample Planning
- Define the Population
o Students, disabled persons, elderly and adult samples report similar levels of
happiness
o When Dieners analyzed surveys of happiness in countries around the world, the
average level of happiness varied markedly (satisfaction with life domains
varies across cultures)
o Cross-population generalizability: compare results obtained from samples of
different populations
- Define Sample Components
o Elements: elementary units, individual member of population whose
characteristics are to be measured
o Sampling frame: List of all elements in a population
o Population: Entire set of individuals or other entities to study
o Representative sample: “looks like” population from which it was selected (in
unrepresentative sample, some characteristics are overrepresented or
underrepresented)
o Sample generalizability depends on the amount of sampling error (difference
between sample and population)
o Estimating Sampling Error
Inferential Statistics: tool for calculating sampling error
Sampling distributions for many statistics have a normal shape
Random Sampling Error: Variation owing purely to chance
Sample Statistic: value of a statistic such as a mean, computed from
sample data PSYB01 Chapter 5 – Sampling and Survey Research
Population Parameter: Value of a statistic computed using the data for
an entire population. Sample statistic is an estimate of a population
parameter
Sampling Methods
- Probability Sampling Methods: Probability of selection is known and is not zero
- Non-probability Sampling Methods: Sampling methods that do not let us know in
advance the likelihood of selecting each element
- Probability of Selection: likelihood that an element will be selected from the population
for inclusion in the sample
Probability Sampling Methods
- These methods have no systematic bias (nothing but chance determines which
elements are included in the sample)
- Four most common methods for drawing random samples
o Simple Random Sampling: procedure where cases are identified on basis of
chance
Random digit dialing: machine dials random numbers within the phone
prefixes
o Systematic Random Sampling: variant of simple random sampling
First element is selected randomly and then every nth element is
selected
Watch out for periodicity (sequence varies in some regular, periodic
pattern) (e.g. houses counted and the house on northwest side always
chosen because of sampling interval [number of cases from one
sampled case to another])
In that situation, starting point needs to be changed
o Stratified Random Sampling: uses information known about population prior to
sampling to make sampling process more efficient
Ensures appropriate representation of elements across strata
Proportionate Stratified Sampling: ensures that sample is selected so
that the distribution of characteristics in the sample matches the
population
Disproportionate Stratified Sampling: Sampling where characteristics of
sample are disproportionate to population PSYB01 Chapter 5 – Sampling and Survey Research
o Cluster Sampling: Sampling selected in two or more stages, first random
selection then random selection of elements within clusters
Clusters are naturally occurring, mixed aggregate of elements of a
population
Nonprobability Sampling Methods
- Availability Sampling (Convenience sampling): available or easy to find, haphazard,
accidental
- Quota Sampling
o Intended to overcome the flaw of availability sampling: that sample will just
consist of whoever or whatever is available
o Must match percentage of desired population study
o No matter what type of sample you draw, your sample will be less
representative if more potential respondents refuse to participate
Writing Survey Questions
- Avoid confusing phrasing
o Simple direct approach to asking a question minimizes confusion (shorter words
and sentences and keep total number of words to 20 or fewer)
o Avoid “don’t” or “not” in questions, double negatives are confusing
o Avoid double-barrelled questions: asking two questions but allow only one
answer
o Use filter questions: survey question used to identify a subset of respondents
and then asked other questions Avoids skip patterns (Combination of

More
Less
Related notes for PSYB01H3