I (I) Ilissycores Incensusce) ()
-
Psychometrics: field that evaluates if a psychological test is good or
garbage
good 18: psychometrics:
· reliable
Psychological tests:
·psychological
walld ↓ egts
a psychological test compares the
behavior of It people
provide asurement
meaningful Stical
systematically collected
Information edures and
inter-individual differences
ncepts
Intra-individual differences
Types of tests:
. based on
purpose
criterion referenced tests:) are each score with a predetermined cut-off point
e.g. If your height is less 70 m, you are short
horm referenced tests:compure each score with a reference sample and horm
e.g. 18 You are shorter than 50% of your class, you are short
2. based on time
· speed tests
ability tests
IQ is a latent variable
a psychological test usually measures things that are not observable
height is measured in meters, you can't directly observe or measure it
how to define latent variables >
Operational definitions
Steps: observed variable
1. measure a latent variable, like working memory is causally related to
2. define with
it operational definition latent variable
3. Then come up with a test!
We will measure working memory by making people do a task.
we will observe/evaluate their behavior/performance on this task
measurement:
the assignment of numbers to objects
=
event to rules
or
according
challenges
composite scores: adding multiple item scores, not always good
score
sensitivity:the ability of a test to
distungish between units. If there Is
a mood test and you can only say 'I am good' or 'I am depressive; you force people
to make a choice, you cannot distungish your sample so well.
lack of awareness:a psychological measurement is socially/culturally influenced.
In the west, intelligence might be difference than in the east.
participant reactivity:when participants screw us over
.
question characteristics:participants change their behavior to expectancy
1. Social desirability:participants try to impress the researchers. researcher
3.
Malingering:participants take data
? Other blazes and expectations
, Ilisycored
"(II)(i)
Incense.
the
measurement: assignment of numbers to objects or events according to rules
gender military (Q height
nominal ordinal Interval rablo scales
dent, tr V V U 3
now numbers are
to
order W V W assigned psycho-
quan W W logical attributes
absolu W
Individual differences
measurements assume that psychological differenc d can be detected by well-
designed measurement processes
If we measure the 1 of 1000 people, we will
get many different scores
The differencebetween scores this distribution is called
in
variability
Quantifying variability
good psychological research quantifies variability as much as possible
central tendency * A
variance/SD R" is variance of explained by X
shape
central tendency
mean: average score in
a distribution a variance/sD indicates greater distribution
of scores larger variability in a
the variance and SD cannot be below 0,so minimum is o
2 X there to interpret the variance and SD as
=>
is no simple way small or
large
= N
x
T
variance/sp
variance:how much the scores in a distribution deviate from the mean
52-**-xt"
-for
standard deviation:square root of variance
v5 =
"
shape
normal distribution:bell shape
so no
comparing 2 distributions >
mean
covariance:the degree of association between the variability in 2 distributions
>positive: a positive association of a variables if one increases, so will the other
rice versa for negative
covariance
Cxy
=
= =xx-x (ly-ig)
-
Psychometrics: field that evaluates if a psychological test is good or
garbage
good 18: psychometrics:
· reliable
Psychological tests:
·psychological
walld ↓ egts
a psychological test compares the
behavior of It people
provide asurement
meaningful Stical
systematically collected
Information edures and
inter-individual differences
ncepts
Intra-individual differences
Types of tests:
. based on
purpose
criterion referenced tests:) are each score with a predetermined cut-off point
e.g. If your height is less 70 m, you are short
horm referenced tests:compure each score with a reference sample and horm
e.g. 18 You are shorter than 50% of your class, you are short
2. based on time
· speed tests
ability tests
IQ is a latent variable
a psychological test usually measures things that are not observable
height is measured in meters, you can't directly observe or measure it
how to define latent variables >
Operational definitions
Steps: observed variable
1. measure a latent variable, like working memory is causally related to
2. define with
it operational definition latent variable
3. Then come up with a test!
We will measure working memory by making people do a task.
we will observe/evaluate their behavior/performance on this task
measurement:
the assignment of numbers to objects
=
event to rules
or
according
challenges
composite scores: adding multiple item scores, not always good
score
sensitivity:the ability of a test to
distungish between units. If there Is
a mood test and you can only say 'I am good' or 'I am depressive; you force people
to make a choice, you cannot distungish your sample so well.
lack of awareness:a psychological measurement is socially/culturally influenced.
In the west, intelligence might be difference than in the east.
participant reactivity:when participants screw us over
.
question characteristics:participants change their behavior to expectancy
1. Social desirability:participants try to impress the researchers. researcher
3.
Malingering:participants take data
? Other blazes and expectations
, Ilisycored
"(II)(i)
Incense.
the
measurement: assignment of numbers to objects or events according to rules
gender military (Q height
nominal ordinal Interval rablo scales
dent, tr V V U 3
now numbers are
to
order W V W assigned psycho-
quan W W logical attributes
absolu W
Individual differences
measurements assume that psychological differenc d can be detected by well-
designed measurement processes
If we measure the 1 of 1000 people, we will
get many different scores
The differencebetween scores this distribution is called
in
variability
Quantifying variability
good psychological research quantifies variability as much as possible
central tendency * A
variance/SD R" is variance of explained by X
shape
central tendency
mean: average score in
a distribution a variance/sD indicates greater distribution
of scores larger variability in a
the variance and SD cannot be below 0,so minimum is o
2 X there to interpret the variance and SD as
=>
is no simple way small or
large
= N
x
T
variance/sp
variance:how much the scores in a distribution deviate from the mean
52-**-xt"
-for
standard deviation:square root of variance
v5 =
"
shape
normal distribution:bell shape
so no
comparing 2 distributions >
mean
covariance:the degree of association between the variability in 2 distributions
>positive: a positive association of a variables if one increases, so will the other
rice versa for negative
covariance
Cxy
=
= =xx-x (ly-ig)
