Textbook Notes (280,000)
CA (160,000)
UTM (8,000)
PSY (2,000)
Chapter 1

PSY230H5 Chapter Notes - Chapter 1: Standard Score, Standard Deviation, Negative Number

by OC4

Course Code
Ulrich Schimmack

This preview shows pages 1-3. to view the full 11 pages of the document.
PSY230- personality psychology
Lectures 1-4 (chp 1)
Introductory pages
what is personality psych?
- individual differences vs everybody same compare personalities of diff ppl
vs. cognitive (basic cognitive process: memory, attention, etc) and dev psych
(changes in psycho processes over life span)
- healthy and normal differences
vs. abnormal psych (assumes ppl diff, but wants to help em)
- personality and situation affect behaviour
vs. social psych (social influences on psych process)
o focus on individual: idiographic
- one indiv at a time
- understand indiv’s actions from indiv’s characteristics
- eg. Biography of famous ppl
- eg. Self-exploration
- More complete understanding of a SINGLE indiv
- Great for important ppl BUT NOT everyday ppl
- Subjective (influenced by biographer)
- Difficult to test scientifically
- Cannot be generalized to other indiv
o focus on variable: nomothetic
- relation b/w individual’s differences in one variable and another variable
- eg. Self-esteem correlates w life satisfaction (r=0.60)
- Objective (replication)
- Theories can be tested scientifically
- More complete understanding of MANY people
- Useful for everyday ppl
- Boring : no juicy stories of personal life events
- Difficult: need to understand correlations
1. classification
- how do ppl differ from each other
- how these differences related to each other
2. causes
- why ppl differ from each other
3. consequences
- what are the effect of indiv differences
Chapter One
- There is no “zero” in psychology. We cannot make an exact ratio to compare two
people. ex: we cannot say this person is 50% more rebellious than another.

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

PSY230- personality psychology
Lectures 1-4 (chp 1)
- There is no “true zero level. This means that lets say a person takes an
intelligence test and gets zero, we cannot say that the person has zero intelligence
because if the test was easier they would have gotten a higher score. Therefore
there are no zero levels in psychology.
- ‘Absolute amount of a variable for example how intelligent someone is. We
cannot measure the absolute amount of a variable in a person.
- In psychology we can still measure given traits in a person. One way of doing this
is by rank. We can rank people from most rebellious to least rebellious and put
them in 1st, 2nd, 3rd place.
- The issue is that the person in first place may only be slightly more rebellious
than the person in 2nd place, and the 2nd place may be much more rebellious than
3rd place therefore it is hard to determine an average. The differences between
each place or “score must be MEANINGFUL. Therefore each difference must be
the same (equal intervals)
Equal diff / intervals between scores represent roughly equal diff in the level of the
Eg) diff b/w IQ 120 and 130 has same meaningful diff as diff b/w 140 and 150.
Standard scores enables to make meaningful comparisons across diff kinds of
measurement scales
Eg) how far below or above avrg on for one trait vs other
- The measurements are not taken in a specific unit. We must come up with a way
to compare two different variables. For example, if your IQ is 90 and the average
is 100, and your average sociability score is 60 and the average is 50, we must be
able to understand that although 60 is lower than 90, the sociability score is
HIGHER than the IQ. These two scores are on different scales.
- Basically in order to compare two scores, you have to change both scores into
STANDARD SCORES so that you can compare properly.
- Take the persons score and subtract the AVERAGE SCORE from it. If the
outcome is a positive number, they are above average, if it is a negative number,
they are below average.
But must also take variablitiy into consideration. So devide diff from person’s score to
avg score by st dev.
- After making the comparison we want to see how FAR a person is from the
mean. Are they WAY above or below average, or not too far off?
- Take the sum from the first step and divide it by the standard deviation.
- The standard score is a universal value.
Standard score mean = 0, st dev = 1
Correlation coef
- Known by the symbolR”
- Values range from -1 to +1. -1<r<+1
- No correlation = 0.
- correlate two variables to each other
- variable: quality that varies quantitatively between diff entities
- indicates strength & direction of the relationship

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

PSY230- personality psychology
Lectures 1-4 (chp 1)
- under 2 is weak. 2-5 is good. 5+ is strong.
- To get an exact meaning of a correlation we use the BESD BINOMIAL
EFFECT SIZE DISPLAY. It tells you how strong the correlation is.
- BESD: 4 cells, each cell represents one combo when comparing two variables.
- We can use the BESD to measure correlation by
Subtracting top left or bottom right from top right or
bottom left.
- statistical sig: p<0.5 . thus, correlation greater than chance: tells us nothing about
the importance of a research finding and it does not tell us how strong the
correlation between two variables is.
- sample size matters: In sum, correlations can vary from group to group if groups
differ in the variability of causal factors. For this reason it is necessary to conduct
studies in different groups or to specify the group in which a correlation was
- A linear prediction means that we use a straight line to predict values of one
variable from values of another variable.
- Higher numbers indicate closer relations between two variables where
observations are close to the regression line. (vice versa)
Eg) correlation of +1 means perfect correlation b/w two variables
So if a person is 2 st dev above mean for height, they would be 2 st dev above
mean for intelligence
Effect size
- reflects strength of an effect independent of sample size
- Explained variance (r2)
- Binominal Effect Size Display (.5 + r/2)
- Effect sizes are most important to compare the effects of different variables.
-observed effect sizes underestimate true effects.
Explained variance = r!
- measure of strength of a correlation: 0 to 1
- if variable is influenced by a single casual factor, variation in the casual factor is a
perfect predictor of variation in the outcome variable
- or how well one variable predicts another variable without making any
assumptions about causalityjust info about strength of the relationship b/w the
two variables
Binomial effect size display (BESD)
- typically overestimates the strength of an effect
- thus use r and r^2
- a correlation of r = .5 implies that we can increase our chances in guessing
whether somebody is above or below average in weight from 50% correct guesses
(without information we will be right 50% of the time and wrong 50% of the
time) to 75% correct guesses (.5 + .5 / 2) if we receive information about height
" it looks more impressive to say that you have a 75% chance of being correct
than to say that you actually just explained 25% of the variance.
- 0.5+r/2
You're Reading a Preview

Unlock to view full version