Assessing Subjective Well-being in Wide Populations. A Posetic Approach. . .
249
A quite complete presentation of the basic definitions and main elements of this
methodology can be found in Fattore 2016 and 2017.
Using poset 10 , it is possible to analyze the different levels of a multidimensional
concept, making a synthesis by variables. It is also possible to compare different
sub-populations of respondents on providing summary measures for each subgroup.
This paper takes the basic elements of the methodology for granted.
3 Data
Eu-SILC is an European harmonized official statistical survey 11 , carried out since
2004 and structured in a longitudinal and a transversal component. It periodically
adopts ad-hoc modules that examine in depth topics of particular importance. The
2013 ad-hoc module measures the state of well-being from a subjective perspective,
and consists of 22 questions, on overall LS, satisfaction with specific aspects
of life (including interpersonal relationships), MoL, emotional states (including
happiness), trust, physical security, availability of someone to talk to and someone to
ask for help. The survey units are households and family members over 16-years old.
The Italian survey collected 38,039 respondents in 2013. Within the subset of people
who answered to the ad-hoc well-being module (25,500 respondents), we select
those between 26 and 65 years old (15,354 records). We have chosen to represent the
three dimensions of SWB through the LS the MoL and the five affects. We have also
selected other variables that we consider important to study the differences in SWB
levels in different subgroups: labour status and gender. In particular, concerning the
labour status (employed, unemployed, retired, etc.) we used the one indicated by the
respondent. In fact, we consider that the self-attributed status is more significant in
defining a relationship with the perceived level of well-being. Naturally the variable
status is categorical non-sortable. 12
10 There is a large literature on the treatment and synthesis of multidimensional systems of ordinal
data using non-aggregative methods, allowing the construction of synthetic measures without the
aggregation of the scores of basic indicators. Within this approach, poset has become a reference
over the years, as demonstrated by many works in different fields of research (for instance, see:
Annoni and Bruggemann 2009; Fattore et al. 2015; Carlsen and Bruggemann 2017; Arcagni et al.
2019). However, poset can also be suitable for quantitative data (see: Fattore 2018; Alaimo 2020;
Alaimo et al. 2020a, b, c), allowing the overcoming of some limitations of the aggregative methods.
11 EC Regulation n.1177/2003.
12 Full-time employees (EFT), Part-time employees (EPT), Full-time self-employed (SEFT), Parttime self-employed (SEPT), Unemployed (UNE), Students (STU), Retired (RET), Unfit to work
(UNF), Fulfilling domestic care (HOU), Other inactive (INA).
249
A quite complete presentation of the basic definitions and main elements of this
methodology can be found in Fattore 2016 and 2017.
Using poset 10 , it is possible to analyze the different levels of a multidimensional
concept, making a synthesis by variables. It is also possible to compare different
sub-populations of respondents on providing summary measures for each subgroup.
This paper takes the basic elements of the methodology for granted.
3 Data
Eu-SILC is an European harmonized official statistical survey 11 , carried out since
2004 and structured in a longitudinal and a transversal component. It periodically
adopts ad-hoc modules that examine in depth topics of particular importance. The
2013 ad-hoc module measures the state of well-being from a subjective perspective,
and consists of 22 questions, on overall LS, satisfaction with specific aspects
of life (including interpersonal relationships), MoL, emotional states (including
happiness), trust, physical security, availability of someone to talk to and someone to
ask for help. The survey units are households and family members over 16-years old.
The Italian survey collected 38,039 respondents in 2013. Within the subset of people
who answered to the ad-hoc well-being module (25,500 respondents), we select
those between 26 and 65 years old (15,354 records). We have chosen to represent the
three dimensions of SWB through the LS the MoL and the five affects. We have also
selected other variables that we consider important to study the differences in SWB
levels in different subgroups: labour status and gender. In particular, concerning the
labour status (employed, unemployed, retired, etc.) we used the one indicated by the
respondent. In fact, we consider that the self-attributed status is more significant in
defining a relationship with the perceived level of well-being. Naturally the variable
status is categorical non-sortable. 12
10 There is a large literature on the treatment and synthesis of multidimensional systems of ordinal
data using non-aggregative methods, allowing the construction of synthetic measures without the
aggregation of the scores of basic indicators. Within this approach, poset has become a reference
over the years, as demonstrated by many works in different fields of research (for instance, see:
Annoni and Bruggemann 2009; Fattore et al. 2015; Carlsen and Bruggemann 2017; Arcagni et al.
2019). However, poset can also be suitable for quantitative data (see: Fattore 2018; Alaimo 2020;
Alaimo et al. 2020a, b, c), allowing the overcoming of some limitations of the aggregative methods.
11 EC Regulation n.1177/2003.
12 Full-time employees (EFT), Part-time employees (EPT), Full-time self-employed (SEFT), Parttime self-employed (SEPT), Unemployed (UNE), Students (STU), Retired (RET), Unfit to work
(UNF), Fulfilling domestic care (HOU), Other inactive (INA).