Study Guides (390,000)
CA (150,000)
UTSG (10,000)
PSY (800)
Study Guide

[PSY230H1] - Final Exam Guide - Everything you need to know! (28 pages long)


Department
Psychology
Course Code
PSY230H1
Professor
Dan Dolderman
Study Guide
Final

This preview shows pages 1-3. to view the full 28 pages of the document.
UTSG
PSY230H1
FINAL EXAM
STUDY GUIDE

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Methodology
September-21-16
9:53 AM
Until you make the unconscious conscious, it will direct your life, and you will call it fate -- C. Jung
Locus of control scale (the second one) [LDS = locus de control]
o Measures the extent of You have control of things vs. what you can't control
o It's not a pure construct -- it's multidimensional
o If control is internal probably means you're more confident, have goals
Is it control or self-esteem? Optimism? Certainty?
o It's a useful tool to measure personality
o Researchers try to refine things
Eg. some may find SE is the same as optimism or they are different/just have some
similarities
But they are unique constructs
Much of personality research is about method
o Researchers create scales and get people to fill them out
o There more than 1 question, so they can see how consistent people answer (to get patterns
from them)
To get a better understanding of the person's personality as a whole
What do you gain by asking similar questions?
To get a holistic understanding
Some people may respond in similar questions differently because of idiosyncrasies
they have
o When we have a scale, we first start to find inter-reliability of it
Who hope that the questions are related to the construct we are measuring
We want to see consistency in people's responses
We can cut the scale in half, and get the mean of both halves and see if they are
correlated with lots of people
The correlation can be driven by outliers? If you split the scale in different ways,
you may get a different correlation
We split the scale every way it can be split, calculate the correlation, and look at
the average of them
This is called chronback's alpha
High CA = all points to the same construct (there is convergence among
the items)
Low CA = the points don't correlate as well (some items are pointing
towards the construct, but the others don't)
Are there other factors in this whole construct if some points don't
point towards it
o Locus control are moderately reliable
Some items hang well
But some points don't, but when you cluster these points, they hang very well
o Researchers look at how well points correlate with each other and the first thing they focus on
is the scale's inter-reliability
o Some scales are wish-washy where some answers are on the extreme
Eg. tests -- we don't want all the questions to be medium, we want to have some easy
and hard questions too
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

This gives you the ability to see the difference between people/you are able to
discriminate b/w people
Eg. "all in all, I am inclined to say I'm a failure"
You're only seeing the extreme here
Eg. "I wish I can have more respect for myself"
A softer question -- medium question
We need multiple choice questions to see the idiosyncrasies of people
Problems with self-esteem scales
o Unstable
People's mind changes all the time, so their answers can vary depending on when they
took the scale
Is this a problem of the measure? Or are you measuring an unstable construct?
Self-esteem is based on deep personal feedback/bedrock psyche = stable construct
If this is the case, then the measure is not great
But maybe self-esteem is unstable
o Scale might be too general
Maybe we are not actually getting self-esteem
Eg. we don't get asked "how do you feel when you fail?"
Maybe we have to ask questions on a particular way to get our construct
This then becomes a validity debate
Reliability
o Internal reliability
How much your scale hangs together/there is convergence
o Test re-test reliability
Does the self-esteem scores of a person similar when it was taken in time 1 and 2
If the answers don't converge, then you have an unreliable measure
This is a big problem in personality research because constructs can be unstable (it can
change in different times)
There is a lot of variability
The debate goes back and forth in regards to personality (is there really is a
personality?)
Walter Michel argues that there is no personality because of great
variability
But others disagree
You can have the effect of context (your personality can change
depending on the context you are in)
You should be able to get the interaction between the person and the
situation and see how your construct behave in different situations
Personality researchers try to deal with the unstable measures -- but this is not
really a problem that we can get rid of
Validity
o What constructs do these items (on the scale) represent?
o Eg. are you measuring self-esteem or something else?
The scale might be measuring depression
People might respond to these questions differently because of culture
Did you just pick up social norms
o To solve validity problem, you do the same thing as you do when you need reliability
Make sure that all items point to something/construct
You have to have reliability in order to have validity
Your points have to point at something (reliability) and you have to know what
that something is (validity)
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version