false

Study Guides
(248,518)

Canada
(121,606)

University of Guelph
(7,161)

Psychology
(952)

PSYC 2360
(37)

Naseem Al- Aidroos
(10)

Midterm

Unlock Document

Psychology

PSYC 2360

Naseem Al- Aidroos

Winter

Description

Things to Remember for Methods Midterm 2
Surveys provide snapshot view of attitudes, behaviours and beliefs.
Interviews: structured (follow a decision tree) or unstructured (focus
group, need to be careful of bias)
Questionnaires: fixed format, without supervision, cheaper
Can increase the number of people that respond to your questionnaire
(response rate) by making it seem brief and interesting, stressing
importance of participation, offering gifts or ensuring confidentiality
Representative Sampling: when sample is the same as the population in all-
important aspects
Probability Sampling: when everyone in the population has a known
chance of being chosen to be a part of the sample
Simple Random Sampling: when sampling frame is obtained, number
from 1 to however many people there are and then use random
number generator to see who is included in sample
Systematic Random Sampling: when sampling frame is already in
th
random order, we choose a number and every n person is included
in the sample
Stratified Random Sampling: when the population is split into
subgroups (stratas) and sampling occurs from the stratas
o Proportionate Stratified Random Sampling
o Disproportionate Stratified Random Sampling
(Oversampling): used when subgroup is small, over sample the
number of people by sampling more and then weigh their
scores less in the measure therefore it is still representative
Cluster Sampling: the population is split into clusters and the
clusters themselves are then chosen for the sample
The larger the sample size, the more likely it will be representative
2 Conditions: (a) there must be at least 1 sampling frame of everyone in
population of interest and (b) everyone chosen to be a part of the sample
must complete the experiment. If either is not met, the sample is
susceptible to sampling bias when it is not representative of the
population Non Probability Sampling: often used when sampling frames cannot be
obtained
Snowball Sampling: sample one individual and then as for a referral
to another individual
Convenience Sampling: sample whoever is available - what we do at
the University of Guelph Psychology Dept.
Picturing the Data: frequency distribution, histogram, stem and leaf plot
Central Tendency: mean (5% trimmed mean), median, mode (can be used
for nominal data)
Dispersion (spread of data): range, interquartile range (25thto 75th
percentiles), and standard deviation (larger it is, the bigger the spread)
Unacknowledged Participant: undercover, ethical issues, emotions can
distort data
Acknowledged Participant: need to be careful of reactivity
Unacknowledged Observer (least reactive) and Acknowledged Observer:
observe within reasonable privacy
Case Studies: can provide ideas for future research, learn about cases
that are unethical to examine experimentally, very limited because person
is normally unique or unusual in their behaviour therefore hard to
generalize
Coding methods should be decided beforehand otherwise you are more
likely to see things that confirm your hypothesis rather than disconfirm.
Event Frequency: how many
Event Duration: how long
Event Sampling: focus on one behaviour for entire group
Individual Sampling: focus on one individual and all of their
behaviour
Time Sampling: focus on one individual for a certain amount of
time and then switch
The null hypothesis is that there is no correlation, no difference; it is the
least interesting outcome. The alternative hypothesis is the researcher’s hypothesis and is what we turn to if we reject the null hypothesis if the
null hypothesis is likely to occur less than 5% of the time.
Type 1 Error (alpha): probability of rejecting the null when the null is
actually true, also known as significance level and is usually set to .05
Type 2 Error (beta): probability of accepting the null when the null is
actually false
Power (1-beta): probability of rejecting the null when the null is actually
false
We can decrease type 1 and type 2 errors by increasing power - the
larger sample size you have the more likely you will find significant
relationships, stronger manipulations cause this as well (two groups
are more different from each other therefore more likely to find
significant differences)
Power is low in social sciences (~.6)
Sampling distributions is when we take an infinite number of samples of
sample size ‘n’ and we calculate a mean for each, we then plot the
frequency of the means to get a sampling distribution of the mean.
Central Limit Theorem: (a) the sampling distribution of the mean is equal
to the population mean (b) as n increases, the variability of the sampling
distribution of the mean decreases and (c) the sampling distribution of
the mean is normal regardless of the underlying population as long as n >
30.
Effect size tells you the magnitude of the relationship (while p

More
Less
Related notes for PSYC 2360

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.