false

Textbook Notes
(368,795)

Canada
(162,165)

University of Toronto Scarborough
(18,529)

Psychology
(9,697)

PSYB01H3
(581)

Connie Boudens
(81)

Chapter 5

Unlock Document

Psychology

PSYB01H3

Connie Boudens

Fall

Description

Chapter 5: Sampling and Survey Research
census : studying the entire population of interest; difficult to conduct if population is too
big
Sample Planning
- Define the population
cross-population generalizability: when findings from one group/population holds true
for another group/population. a.k.a external validity; to assess this, you need to compare
results of groups.
- ex: study found differences in how Eastern and Western cultures perceived
identical scenes.
clearly define the population in question, and not give the impression that data came from
a different group
-Define sample components
elements: individual members of population. ex. students
sampling frame: list of all the elements in population. ex: list of students from the
registrar’s office
population: entire set of individuals to which the findings are to be generalized
representative sample: sample that looks like that entire population; it has the same
distribution of characteristics and does not over represent or under represent the
population
sampling generalizability depends on the amount of sampling error: diff between
characteristics of the sample and of the entire population it came from.
- larger error = less representative of the sample = less generalizable findings
Estimating Sampling Error
- inferential statistics: mathematical tool that estimates how likely a statistical
result from a random sample represents the population it came from.
- think of our sample as one possible sample out of many that could have been
chosen
- sampling distribution: a plot of the mean values of all the samples
- a normal distribution is symmetric; due to random sampling error: error plainly
due to chance and not to systematic errors.
- sample statistic: the statistic computed ex: mean; it is an estimate of a population
parameter
-population parameter: value of statistic ( ie. mean) from the entire population
- sample statistic and population parameter should be identical; peak of the bell
curve represents the norm for the entire population
- in a normal distribution, a predictable proportion will fall within certain ranges
Sample Methods
Probability sample methods: method that allows us to know how likely an element will
be chosen (probability of selection is known)
- relies on chance. ex: flipping a coin = ½ or 50% chance → probability of
selection
- no systematic bias Four Most Common Methods of Random Sampling:
- Simple Random Sampling
→ strictly on the basis of chance
→ probability of selection is equal for each element
→ can be computer generated ex: random-digit dialing done by organization doing
phone surveys
- Systematic Random Sampling
→ variant of simple random sampling
→ first element is selected randomly from a list and then every nth element is
selected.
→ convenient for drawing samples from sequentially arranged population elements
ex: folders in a filing cabinet
→ periodicity: sequence varies in some regular periodic pattern
- Stratified random Sampling
→ uses information known about the total population prior to sampling to make the
process more efficient
→ first, all elements are distinguished according to some relevant characteristic ex:
racial background → makes the sampling strata
→ second, elements are sampled from within these strata. ex: people are chosen from
the racial background
→ size of each stratum must be known
→ this method assure appropriate representation of elements across strata
→ proportionate stratified sampling: distribution of characteristics within a selected
sample matches the distribution of entire popu. ex: population ratio of undergrads to
graduate students is 2:1 therefore # of students in both strata should replicate the ratio
→ disproportionate stratified sampling: ex. Eisenberg mental health study where
he chose a larger portion of graduate students versus undergrads because graduate
students’ mental health were “understudied”; able to calculate separate statistical
estimates.
- Cluster Sampling
→ useful when sampling frame is unavailable ie larger populations spread across wide
area
→ ex: schools are clusters for sampling students, blocks are clusters for sampling city
residents
→ list of clusters is easier to obtain than a list of each person in that cluster ex: list of
schools is easier than a list of students in all of Toronto
→ researcher draws random samples from each cluster
→ cluster and stratified sampling are usually used in a combo
Non probability Sampling Methods: does not use random selection
-ex: availability sampling: elements selected because they are easy to find or available.
- Quota Sampling
→ quotas are set to ensure that the sample represents certain characteristics in
proportion to the population
→ ex: in a town, 48% are male. You want a sample of 500 residents, therefore you
look for 240 males (48% of 500). The percentages are you quotas.
→ differs from stratified sampling in that quota does not need a sampling frame and is
not random.
→ problem of this method is the potential of respondents not wanting to participate ex:
phone surveys Writing Survey Questions
Avoid Confusing Phrasing
- be simple and direct
- shorter words and shorter sentences; 20 or less than words
- avoid double negatives:
→ Do you disagree that rich people are not happy?
- avoid double-barreled questions: asking two questions in one is confusing
→ Agree or Disagree: Cars should be faster and safer. → can agree to one
but disagree to another
- make sure that respondents can relate to questions using filter questions and skip
patterns.
→ Are you employed? If NO, skip to question 16.
Minimize the Risk of Bias
- answers can be biased due to the phrasing
Avoid Making Either Disagreement or Agreement Disagreeable
- when dealing with disapproved behaviour, respondents tend to be reluctant t

More
Less
Related notes for PSYB01H3

Join OneClass

Access over 10 million pages of study

documents for 1.3 million courses.

Sign up

Join to view

Continue

Continue
OR

By registering, I agree to the
Terms
and
Privacy Policies

Already have an account?
Log in

Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.