3 RESEARCH METHODS
3.1 Measurement Model
Immigrant well-being is the basic material needs,
health, social relationships, security and freedom of
choice and movement to maintain a high quality of
life. Amartya Sen pointed out that well-being consists
of a set of feasible capabilities, which reflects the
combination of functional activities that an individual
possesses, can achieve, and can freely choose. The
connotation of well-being has multiple attributes such
as complexity, multi-dimensionality, and hierarchy.
It is a complex task to generally discuss the groups of
factors that affect well-being, and the evaluation
model needs to take into account the multi-
dimensional indicators (Walter 2016). Ordinary
econometric simultaneous equation models can only
deal with complex relationships between measurable
variables, while structural equation models (SEM)
can not only deal with the mutual causal connections
between measurable variables, but also use factor
analysis to construct multiple "" Latent Variables”,
and discusses the complex relationship between
latent variables or between latent variables and
measurable variables (observed variables), this model
provides measurement possibilities for variables that
are highly abstract and difficult to measure (Li 2018,
Hu 2020). In view of this, this paper chooses
structural equation modelling as the research method
to explore the complex factors affecting the well-
being of immigrants.
Structural equation model consists of two parts:
measurement model and structural model. The
measurement model is used to describe the
relationship between latent variables and observed
variables, and the structural model is used to describe
the relationship between latent variables and consider
the influence of variation and error terms.
The measurement structure model is expressed as:
X=Λ
x
ζ+δ
Y=Λ
y
η+ε
In the formula, X and Y are exogenous explicit
variables and endogenous explicit variables,
respectively;
δandεare the error terms of X and Y
variables, respectively;
ζare external latent variables;
ηare internal latent variables; Λ
x
factor loadings
for
ζpairs of X variables; Λ
y
actor loadings forη pairs
of Y variables.
Structural models can describe the relationship
between latent variables and can take into account the
effects of variation and error terms. The matrix
equation for the structural model is expressed as:
η=βη+Γξ+ζ
In the formula,
η is the internal latent variable; ξ
is the external latent variable; ζ is the residual of the
internal latent variable;
β is the structure coefficient
matrix between the internal latent variables;
Γ is the
structure between the external latent variable and the
internal latent variable coefficient matrix.
3.2 Selection of Indicators
As involuntary resettlement of large and medium-
sized reservoirs, they face social remodelling and
other problems after relocation. The resettles in the
new environment are vulnerable to new economic
and social risks, which further lead to psychological
risks. The government plays an important role in the
involuntary resettlement process, and the government
and government policies have always had a major
impact on the well-being of immigrants. At the same
time. The selected indicators refer to the monitoring
and evaluation index system for the late support of
large and medium-sized reservoir migration
published in the Notice on Strengthening the
Supervision and Inspection of the Implementation of
the Late Support Policies for Reservoir Migration
published by the Ministry of Water Resources,
including the release of funds, project
implementation, the guarantee system for the
implementation of the late support policies, and the
implementation effect of the late support (Li 2017).
Therefore, the selection of immigrant well-being
indicators needs to consider various factors such as
policy, economy, society, health, family,
environment, psychology, and external well-being.
Based on the perspective of sustainable development,
this study selects a number of indicators that can
reflect the functional areas of immigrant life, and
characterizes the influencing factors of immigrant
well-being in multiple levels and dimensions.
(1) Post-support policies: including fund
distribution, project implementation, guarantee
system for post-support policy implementation, and
post-support policy implementation effects.
(2) Economic and living standards: including
income, consumption, housing, cultivated land,
education, medical care, transportation, drinking
water. There are seven objective indicators were
selected: annual net income, annual total expenditure,
per capita housing area, per capita arable land area,
education level, toilet type and cooking energy use
mode of the interviewed immigrant households; at the
same time, 4 subjective indicators of satisfaction with
education, medical care, transportation and drinking