Study Guides (390,000)
CA (150,000)
York (10,000)
PSYC (1,000)

PSYC 2030 Study Guide - Midterm Guide: Face Validity, Discriminant, Stratified Sampling

Course Code
PSYC 2030
Rebecca Jubis
Study Guide

This preview shows page 1. to view the full 5 pages of the document.
PSYCH2030 Lecture 7 13-03-10 9:07 PM
Sorry for missing Lecture 6, I was ill and not able to make it to class.
*Multiple varied analysis will be a multiple choice question*
Test 2: Next Week
Observational and Survey Research Methods
Goals of Psychology:
1. Describe
2. Predict
3. Control
4. Explain Behaviour
Observational Research:
1. Purpose is to describe behaviour as it naturally occurs in a given
2. Two Types:
o Naturalistic Observations
Absence of Control
Unable to Infer Causality
Observer Bias
Ways of Reducing Observer Bias:
Having multiple observers and assessing the
inter-observer reliability
*Behaviour Checklist (also operationalize
your terms)
o Participant Observations
Use different sampling procedures
1. Time Sampling
2. Event
3. Individual
To reduce Subject Reactivity:
1. use unobtrusive measures
You're Reading a Preview

Unlock to view full version

Only page 1 are available for preview. Some parts have been intentionally blurred.

o direct unobtrusive measures
o indirect unobtrusive measures
Survey Research- a structured set of questions are given to measure
people’s attitudes/beliefs.
Two Types of Sampling:
1. Probability Sampling
o Population and sample
2. Non-Probability Sampling
In order to generalize from your sample to the population, your sample must
be representative of the population (it must be unbiased).
Probability Sampling
Type of Research Sampling:
1. Random Sampling
2. Cluster Sampling- you randomly select a cluster of people all having
a common feature randomly selected.
i. i.e. select 7 out of 20 buildings
ii. select 3 out of 25 floors
iii. select 10 out of 50 people from those floors.
iv. Therefore, rather than 20x25x50=25 000, it is 7x3x10=210
which is more realistic and simple yet random.
3. Stratified Sampling- is where proportions of important subgroups in
the population are exactly represented in your sample
2,000 ~20yrs olds
500 ~50yrs old
Total: 2,500
So you randomly pick 100 but proportions must remain:
80 ~20yrs old
20 ~50yrs old
Total: 100 people
Non-Probability Sampling (used in experiments)
You're Reading a Preview

Unlock to view full version