Baltatescu

Comparative results and psychometric proprieties of the
Personal Wellbeing Index Romania (old and new

versions) with an adolescent sample
- a preliminary analysis
Discussion document 2006
Sergiu Bltescu
University of Oradea
www.sergiubaltatescu.info
A. Context
Among the European countries, Romania is one of the most religious. Even if is a secular
state, only 0.1% of the citizens declare themselves to be of no religion and/or atheist (data
based on the 2002 census). Almost all of the citizens are Christians, affiliated with the
Romanian Orthodox Church (86.7%), Roman Catholicism (4.7%), Protestantism (3.7%),
Pentecostal denominations (1.5%) and the Romanian Greek-Catholic Church (0.9%).
The school system is secular; however, following the religious revival after the fall of the
communism, optional religious courses were introduced. The young are even more
religious than the average Romanian. Church attendance is high for this group, as the
next table shows:
Table 1. Church attendance in the last month of the urban school children (8th-12th classes).
%
Never
once or twice
three times or more
42,6
25,0
32,4
The main research question here is: how relevant is the insertion of a new question about
satisfaction with religion/spirituality?
B. Method
The questions discussed here are part of the Adolescents future citizens. A
longitudinal study of the process of social exclusion and integration of the school
adolescents financed by CNCSIS (research grant A 497/2006). The questionnaire also
includes various measures of psychological and subjective well-being, social exclusion
and victimization from the peer group, social competency and leadership, school climate,
civic and political involvement and attitudes.
A representative sample 3509 school adolescents (from 8th to 12th classes, 97% of them
between 14 and 18 years old) participated in the survey. The respondents (boys 45.2%,
girls 54.8%) filed the questionnaire in the classes, under the supervision of a trained
Electronic copy available at: http://ssrn.com/abstract=2547193
research assistant. Full confidentiality was guaranteed for the respondents. The research
was completed between November 2006 and February 2007.
C. Values
Overall, the respondents show high levels of subjective well/being.
Table 2. Average values for PWI (old and new versions) and life as a whole.
Class
Personal wellbeing Index (old
version)
Personal wellbeing Index (new
version)
Life as a whole
th
th
10
th
11
th
12
th
TOTAL
82,1
80,3
79,0
77,0
75,7
79,9
82,7
80,0
81,2
76,6
79,3
76,6
77,1
73,7
75,6
72,9
80,3
77,2
As we can see, the ratings computed with the second version are slightly improved. As in
other surveys from Romania, the satisfaction with life as a whole is lower than PWI.
All subjective wellbeing measures decrease with the class (that is, with age), and this
suggest an interpretation of the fact that the average SWB levels in Romania are rather
low: these low scores are the results of the adverse life conditions that accumulate in the
adulthood.
Table 3. Scores on PWI items.
1. Standard of living
2. Health
3. Achievements in life
4. Personal relationships
5. How safe you feel
6. Community connectedness
7. Future security
8. Religion of spiritual fulfilment
8th
8,3
8,5
8,1
8,3
8,5
8,3
8,3
8,3
9th
8,1
8,3
7,8
8,2
8,3
8,1
8,0
8,4
10th
7,7
8,2
7,8
8,0
8,2
7,9
8,0
8,0
11th
7,5
8,0
7,6
7,8
8,1
7,7
7,6
7,8
12th
7,2
7,9
7,6
7,7
7,8
7,7
7,5
7,5
TOTAL
7,9
8,3
7,9
8,1
8,3
8,1
8,0
8,1
As we can see from the above table, the decreasing of the ratings with age is similar for
all indicators.
C. External validity
Both versions of the indexes are almost perfectly correlated.
Table 4. Bivariate correlations between different SWB measures
Life as a
whole
Life as a whole
Personal Wellbeing Index (old)
Personal
Wellbeing
Index
(old)
0,60
1
Personal
Wellbeing
Index
(new)
0,61
0,98
Happiness
(1-4 scale)
0,45
0,40
Electronic copy available at: http://ssrn.com/abstract=2547193
Personal Wellbeing Index (new)

Happiness (1-4 scale)
0,40
1
However, the correlations of the PWI with life as a whole and happiness are rather
moderate.
D. Realibility
1. Factorial structure of both indexes
Both indexes variables were introduced together in a Principal Component Analysis.

Results were rotated using Varimax method with Kaiser Normalization. A single factor
emerged in both cases, explaining over 54% of the total variance. The loadings for both
indexes are shown in the next table.
Table 5. Single factor loadings of the PWI items and life as a whole
Life as a whole
Standard of living
Health
Achievements in life
Personal relationships
Personal safety
Community connectedness
Future security
Religion or spirituality
Total variance explained
Personal WB
(old version)
Personal WB
(new version)
0,71
0,71
0,68
0,78
0,77
0,80
0,79
0,77
56,7%
0,70
0,70
0,68
0,77
0,77
0,80
0,79
0,77
0,61
54,0%
A single factor solution emerged. The loading of the religion/spirituality item is

significantly lower than that of the other items and the insertion of this item slightly
decreases the communality of the single factor.
2. Unique contribution of domains to life as a whole: Multiple regression
I regressed life as a whole against the PWI items, in order to compare the unique
contribution of the domain satisfactions to the global subjective well-being indicator. The
overall fit is moderate. The variations of PWI items explain 41% of the variation of the
dependent variable.
Table 6. Independent standardized contributions of the PWI items to the life as a whole and the
overall fit of the models with the old and the new PWI items as predictors and Life as a whole as a
dependent variable. ** signifies that the beta coefficient is significant at the 0.01 level.
Standard of living
Health
Achievements in life
Personal relationships
Personal safety
Community connectedness
Future security
Religion or spirituality
Adj. r. square
Personal WB (old
version)
Personal WB (new
version)
0,28**
0,05**
0,20**
0,09**
0,01
0,11**
0,07**
0,28**
0,04**
0,20**
0,08**
0,01
0,11**
0,06**
0,05**
41,0%
41,0%
As we can see, the newly introduced item has a low positive contribution to the life as a
whole. On the other hand, it doesnt improve the overall fit of the model. It only steals
little variation from there items (Health, Personal relationships, Future security) that also
have low contributions to the variation of the life as a whole.
Satisfaction with the standard of living, with beta = 0.28 has the largest contribution to
the prediction of life as a whole. Other domains that also predict the dependent variable
are personal achievements (=.20), community connectedness (=.11), personal
relationships (=.09), and future security (=.07). Health (=.05) is significant only
because of the high sample, and personal safety has no independent contribution.
E. Conclusions and limitations

Both versions of the indexes have good psychometric proprieties for the studied group.
All subjective wellbeing measures show that this variable is decreasing with age. As in
other low income countries, satisfaction with the standard of living is the best predictor
for satisfaction with the life as a whole. However, achievements and relationship items
are also significant.
Overall, the insertion of the spirituality/religion item changes very little in the index:
the overall value of the index increases with only 0.5 percent
the fit of the regression model with PWI items as independent variable and life as
a whole as dependent variable does not increase.
the small contribution of the item to the variation of the life as a whole is
significant only because of the high sample.
the item steals explained variance from other low predictors of the life as a
whole.
However, previous studies show that the contribution of the PWI items to the
explanation of the variance of the life as a whole seems to be different in adulthood
compare with that of the adolescents. Therefore, the new PWI should be tested also with
an adult sample.

Baltatescu

Uploaded by

Copyright:

Available Formats

Baltatescu

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Baltatescu

Uploaded by

Copyright:

Available Formats

Comparative results and psychometric proprieties of the

Personal Wellbeing Index Romania (old and new

Electronic copy available at: http://ssrn.com/abstract=2547193

Electronic copy available at: http://ssrn.com/abstract=2547193

Personal Wellbeing Index (new)

Both indexes variables were introduced together in a Principal Component Analysis.

A single factor solution emerged. The loading of the religion/spirituality item is

2. Unique contribution of domains to life as a whole: Multiple regression

E. Conclusions and limitations

You might also like