1.12.2011
A revised theory and instrument to
measure basic
individual values
Shalom H. Schwartz
The Hebrew University of Jerusalem
msshasch@mscc.huji.ac.il
1
1.12.2011
Overview
• Development of a revised theory of basic
values
• The new instrument to operationalize the
theory
• Empirical tests of the new theory: CFA, MDS
• Usefulness of the new theory for explanation
and prediction
2
1.12.2011
Basic Values
Beliefs about the importance of abstract
goals as guiding principles in life
Freedom wealth equality security
pleasure obedience tradition helping
When activated, become infused with
emotion
3
1.12.2011
Security
Circle Organized by Motivational Congruence and Opposition
4
1.12.2011
Security
Circle Organized by Motivational Congruence and Opposition
5
1.12.2011
6
1.12.2011
Dependability
7
1.12.2011
Measuring Values
How much like you is this person?
57 Item Portrait Value
Quest. Examples
Not A little SomeNot
like me like like me what
like me
at all me
Like
me
Very
much
like
me
SD-action: Freedom to choose what he
does is important to him
1
2
3
4
5
6
SD-thought: It is important to her to form
her own opinions and have original
ideas.
1
2
3
4
5
6
UN-tolerance: It is important to him to
listen to people who are different from
him.
1
2
3
4
5
6
UN-nature: He strongly believes that he
should care for nature.
1
2
3
4
5
6
UN-concern: Protecting society’s weak and
vulnerable members is important to
him.
1
2
3
4
5
6
8
1.12.2011
Measuring Values Exemplary Items 2
BE—dependability It is important to him to be a dependable and
trustworthy friend.
BE—caring
It's very important to him to help the people dear to him.
Humility
It is important to him to be humble.
CO—Interpersonal It is important to him to avoid upsetting other people.
CO –rules
Obeying all the laws is important to him.
Tradition
He strongly values the traditional practices of his culture.
SE—societal
Having order and stability in society is important to him.
SE—personal
He avoids anything that might endanger his safety.
9
1.12.2011
Measuring Values Exemplary Items 3
Face
Protecting his public image is important to him.
PO—resources
Being wealthy is important to him.
PO—dominance
He wants people to do what he says.
Achievement
Being very successful is important to him.
Hedonism
Enjoying life’s pleasures is important to him.
Stimulation
Excitement in life is important to him.
10
1.12.2011
Interim Summary & Questions
• 19 values for better prediction & explanation
• Finer partitioning truer to idea of continuum
• Ordered by same principles
• PVQ-R measures each with 3 items
• Can all 19 be distinguished?
• Are they ordered as theorized & collapsible to
10?
• Are the distinctions useful?
11
1.12.2011
Country
Samples
Database
N
Method
Researcher
Finland
Adult
334
P&P 6pt.
M. Verkasalo & K. Porkka
Germany
Student
325
P&P 6pt.
C. Beierlein
Israel
Student
394
Online 6pt
Y. Cohen & S. Schwartz
Italy
2: Adult &
Student mix
388
382
P&P 11pt
P&P 6pt
M. Vecchione
New Zealand
2: Student
141
527
Online 6pt
Online 11pt
R. Fischer
Poland
2: Student &
Adult mix
545 P&P 6pt
1295 P&P 11pt
J. Cieciuch
Portugal
2: Adult &
Student mix
295
297
P&P 6&11pt
P&P 6&11pt
A. Ramos
Switzerland
Student
201
Online 6pt
E. Davidov
Turkey
2: Student
250
240
P&P 6pt
P&P 11pt
K. Demirutku & O. Gumus
USA
Student
443
Online 11pt
M. Konty
12
1.12.2011
Confirmatory Factor Analysis 1
• Combined samples to get large N for CFA
• Gave each sample equal weight in analyses
• One combination of 9 samples with 6pt scale
• One combination of 6 samples with 11pt scale
• Separate analyses in each combined sample
13
1.12.2011
Confirmatory Factor Analysis 2
• Value is latent variable
that causes responses to
items
• If values are distinct,
their items caused only
by one latent value
• Each response also has
some error
• Latent values should be
correlated, but not too
much
14
1.12.2011
Confirmatory Factor Analysis 3
Model
2
df
CFI
SRMR
RMSEA
Initial model (6 pt): 57 items, 19 factors
+1 common
7738
1311
.886
.039
.039
Initial model (11 pt): 57 items, 19
factors +1 common
4837
1311
.890
.040
.031
15
1.12.2011
Confirmatory Factor Analysis 3
Model
2
df
CFI
SRMR
RMSEA
Initial model (6 pt): 57 items, 19 factors
+1 common (n~3680)
7738
1311
.886
.039
.039
Initial model (11 pt): 57 items, 19
factors +1 common (n~2780)
4837
1311
.890
.040
.031
Revised model (6 pt): 48 items, 19
factors +1 common
4615
861
.914
.033
.037
Revised model (11 pt): 48 items, 19
factors +1 common
2987
861
.917
.034
.030
16
1.12.2011
Model Collapsing to Original Values
e1
bed1
e2
bed2
e3
bec1
e4
bec2
e5
bec3
e6
unc1
e7
unc2
e8
unc3
e9
unt1
e10
unt2
e11
unn1
e12
unn2
e13
unn3
BED
BENEVOLENCE
BEC
UNC
UNT
UNIVERSALISM
UNN
17
1.12.2011
Confirmatory Factor Analysis 3
Model
2
df
CFI
SRMR
RMSEA
Initial model (6 pt): 57 items, 19 factors
+1 common
7738
1311
.886
.039
.039
Initial model (11 pt): 57 items, 19
factors +1 common
4837
1311
.890
.040
.031
Revised model (6 pt): 48 items, 19
factors +1 common
4615
861
.914
.033
.037
Revised model (11 pt): 48 items, 19
factors +1 common
2987
861
.917
.034
.030
Second-order model (6 pt): 10 original
basic values
5973
981
.886
.047
.040
Second-order model (11 pt): 10 original
basic values
3962
807
.883
.048
.033
18
1.12.2011
Multidimensional Scaling Analyses (MDS)
•
•
•
•
MDS used for visual assessment of value structure
Each item represented as a point in space
Approach true to assumption of continuum
Shows relations among items, based on similarity in
how respondents rate them (correlations)
19
1.12.2011
Multidimensional Scaling Analyses (MDS)
• Confirmatory MDS asks:
– Do items that represent each value form a spatial region
together?
– Are the regions of values ordered in the space as specified
by the theory based on motivational compatabililty?
12
1
2
11
3
10
4
9
8
7
6
5
20
1.12.2011
MDS combined 9 countries, 6pt
16
19
15
BEC
SDT
BED
ST
SDA
2
1
UNT
HE
3
4
UNC
17
AC
UNN
5
18
HUM
6
14 &
13
POD
COI
12
COR
SES
11
7
POR
SEP
10
9
TR
Face 8
21
1.12.2011
Usefulness of Distinctions
• Is it worthwhile to make all 19 distinctions?
– Does every value have significantly different
associations from all the others?
– Most problematic: pairs of adjacent values
• Adjacent values on continuum—share
motivational goals
– Expect them to have similar associations
– Compare correlations with variables that capture
their distinctiveness
22
1.12.2011
Assessing Usefulness of Distinctions
• Survey included items intended to distinguish
• Compare correlations for pairs of adjacent values
• Do correlations differ:
– significantly in combined samples?
– in expected direction across all samples?
• Two examples per pair of adjacent values
23
1.12.2011
Values Compared # Samples as
predicted
Correlations
Distinguishing Variables
HE
.13
AC
.37
15
Approach Index: In my school/work: I strive to do
well compared to others/ my goal is to perform
better than others
.10
.30
15
Prove Index: In my school/job: I strive to have
others think well of my work/my goal is to have a
good reputation
AC
.37
POD
.31
13
Approach Index
.30
.19
14
Prove Index
POD
.22
POR
.38
14
It's hard to get ahead in life without lots of money
.11
-.03
15
I follow politics closely and form opinions on many
issues.
24
1.12.2011
Values Compared # Samples as
predicted
Correlations
Distinguishing Variables
POR
-.05
FAC
.11
15
It’s terribly wrong to pay cash with no receipt to
avoid taxes
.38
.22
15
It's hard to get ahead in life without lots of money
FAC
.21
SEP
.35
14
The police should have more powers so they can
protect us against crime
.27
.38
14
A person's family should be the main priority in
life
SEP
.18
SES
.30
15
In your life, how important are voluntary
organizations?
.41
.32
15
In choosing job, how important is a high income?
25
1.12.2011
Values Compared # Samples as
predicted
Correlations
Distinguishing Variables
SES
.40
TR
.60
15
In your life, how important is religion?
.14
.05
15
The right to individual freedom is inviolable, it has
to be maintained at all costs
TR
.21
COR
.29
14
It’s terribly wrong to pay cash with no receipt to
avoid taxes
.46
.28
14
Apart from religious services, how often do you
pray?
COR
.28
COI
.12
15
The police should have more powers so they can
protect us against crime
.02
.11
13
We should not try to impose own views on people
who disagree with us
26
1.12.2011
Values Compared # Samples as
predicted
Correlations
Distinguishing Variables
COI
.07
HUM
.25
13
It’s terribly wrong to lie to avoid making person
angry with me
.04
-.09
15
It's hard to get ahead in life without lots of money
HUM
.06
UNN
.22
15
I follow politics closely and form opinions on many
issues.
.29
.07
15
A person's family should be the main priority in
life
UNN
.21
UNC
.30
12/13
.09
.17
15
Immigrants should be given same rights as
everyone else
We should not try to impose our own views on
people who disagree with us
27
1.12.2011
Values Compared # Samples as
predicted
Correlations
Distinguishing Variables
UNC
.23
UNT
.12
14
The government should reduce differences in
income levels
.15
.07
14
A person's family should be the main priority in
life
UNT
.24
BEC
.05
13/13
People who come here from other countries
make this better place to live
.12
.00
15
Homosexuals should have the same rights as
married couples
BEC
-.06
BED
-.17
14
Age: Benevolence—Dependability drops more
with age than Benevolence—Concern does
.28
.20
13
A person's family should be the main priority in
life
28
1.12.2011
Values Compared # Samples
Correlations as predicted
Distinguishing Variables
SDT
.30
SDA
.23
15
Mastery Index: My goal in school/job is: to learn
as much as possible/completely master material
.14
.09
14
I follow politics closely and form opinions on
many issues.
SDA
.27
ST
.19
12
The right to individual freedom is inviolable, it
has to be maintained at all costs
-.15
-.31
14
Age: Stimulation drops more with age than SelfDirection—Action does
ST
.06
HE
.15
15
It's hard to get ahead in life without lots of
money
.07
-.02
15
I follow politics closely and form opinions on
many issues.
29
1.12.2011
Values Compared # Samples as
predicted
Correlations
Distinguishing Variables
BED
.26
SDT
.36
14
In choosing a job, how important would it be that
the job enable you to use your own initiative?
.08
.14
15
I follow politics closely and form opinions on many
issues.
TR
-.31
HUM
-.02
15
Homosexuals should have the same rights as
married couples
.60
.32
15
In your life, how important is religion?
BEC
.08
UNC
.23
15
The government should reduce differences in
income levels
.09
.30
13/13
Immigrants should be given same rights as
everyone else
30
1.12.2011
Benevolence-Care vs. Universalism-Concern
• Benevolence is directed to the welfare of the ingroup
• Universalism is directed to the welfare of those beyond
the ingroup in the wider society
UNC>
Government should reduce differences in income levels
Homosexuals should have the same rights as married couples
Immigrants should be given same rights as everyone else
I follow politics closely & form opinions on many issues
It’s terribly wrong to pay cash with no receipt to avoid taxes
In your life, how important are voluntary organizations?
BEC>
A person's family should be the main priority in life
Some people are simply inferior to other people
In choosing job, how important is a high income?
31
1.12.2011
Conclusions
• Motivational continuum partitionable into 19 values
• Truer to idea of continuum, but still arbitrary
• Order of values same, except BE/UN???
• PVQ-R reliable measure for each, replace 9 items
• Adjacent (& nonadjacent) values have meaningful,
distinctive associations
• Next:
– Finish revision & distribute PVQ-R
– Studies using PVQ-R and theory to predict and explain
significant variables (behaviors, personality, attitudes)
32
1.12.2011
Kiitos huomiosta
33