![Adpated from Zubin and Spring [32]’s vulnerability–stress model in moving home.](https://www.researchgate.net/publication/364761526/figure/fig1/AS:11431281419212076@1746190142624/Adpated-from-Zubin-and-Spring-32s-vulnerability-stress-model-in-moving-home_Q320.jpg)
Access to this full-text is provided by MDPI.
Content available from Urban Science
This content is subject to copyright.
Citation: Cheung, K.-S.; Wong, D.
Measuring the Stress of Moving
Homes: Evidence from the New
Zealand Integrated Data
Infrastructure. Urban Sci. 2022,6, 75.
https://doi.org/10.3390/
urbansci6040075
Academic Editor: Jean-Claude Thill
Received: 10 August 2022
Accepted: 21 October 2022
Published: 25 October 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Measuring the Stress of Moving Homes: Evidence from the
New Zealand Integrated Data Infrastructure
Ka-Shing Cheung 1, *and Daniel Wong 2
1Department of Property, The University of Auckland, Auckland 1010, New Zealand
2Tamaki Regeneration Company, Auckland 1072, New Zealand
*Correspondence: william.cheung@auckland.ac.nz
Abstract:
Moving homes has long been considered stressful, but how stressful is it? This study is
an original attempt to utilise a micro-level individual dataset in the New Zealand Government’s
Integrated Data Infrastructure (IDI) to reconstruct the Social Readjustment Rating Scale (SRRS) and
thereby measure stress at a whole-of-population level. The effects of residential mobility on people’s
mental well-being in the context of their stress-of-moving homes are examined. By using difference-
in-differences analysis, this study scrutinises the stress level across movers, namely homeowners
and renters (i.e., treatment groups) and non-movers (i.e., a control group). The results show that
the change in residence increases people’s overall stress levels. Homeowners are more stressed
than renters, with non-movers as the counterfactuals. Furthermore, the frequency of change in
residences increases individual baseline stress levels. By progressing the understanding of such
stresses, residential mobility researchers can contribute to broader discussions on how individuals’
interpersonal history and social mobility influence their experience. The whole-of-population-based
SRRS will better advance our current ways of measuring mental stress at a population level, which is
crucial to broader discussions of people’s well-being.
Keywords:
residential mobility; well-being; housing tenure; ownership; rental; stress scale; life
course frameworks; New Zealand’s Integrated Data Infrastructure (IDI)
1. Introduction
Moving homes is a fundamental human experience. While residential mobility can
be a mechanism that favourably enables individuals to adjust their housing preferences
and well-being, leaving a familiar neighbourhood and relocating to another geographical
region often has negative impacts. In moving house and home, people must break their
routines and re-establish their social networks [
1
]. Such a transitory process can also cause
much stress and anxiety. Boston Medical Centre suggests that moving house more than two
times per year indicates housing instability that increases the probability of adverse health
outcomes [
2
]. Research in psychology further suggests that individuals from households
that frequently move from place to place, such as military households, have an increased
risk of suicide, substance abuse and even early death [3].
However, at the longitudinal census level, the current understanding of the impact of
moving homes is limited. Some researchers, such as Rumbold et al. [
4
], focus more on the
effects of house moves on children. The reason is that longitudinal studies, particularly
around the perception of the moving experience and the stress endured before and after
changes in residence, often involve large-scale surveys and are very costly to conduct [
5
].
Other medical measurements of stress using cortisol, sex hormones, and blood pressure
could be more objective [
6
] but are still too expensive to be extended into a census scale.
Therefore, an instrument measuring stress objectively without costly sampling is always
preferred. Morris et al. suggest that studies need to take account of life events to assess
potential confounding or be aware of potential biases [
7
]. The limitation of longitudinal
analysis prevents this.
Urban Sci. 2022,6, 75. https://doi.org/10.3390/urbansci6040075 https://www.mdpi.com/journal/urbansci
Urban Sci. 2022,6, 75 2 of 17
This study aims to achieve the following two main objectives:
(1)
Reconstruct a novel measurement of stress in psychology, namely the Social Readjust-
ment Rating Scale, hereafter SRRS for short [
8
], by leveraging a micro-level individual
dataset in the Integrated Data Infrastructure (IDI) of New Zealand [9];
(2)
Demonstrate how to apply this whole-of-population-based SRRS methodology to
conduct future research, to examine and drill down into the stresses associated with
a change in residence for homeowners, renters and non-movers, alongside their
frequency of moving homes, without conducting a single survey or longitudinal study.
This natural experiment of studying the stress of moving home at the whole-of-
population scale is novel for several reasons. First, this is an original attempt to adopt the
SRRS into a census micro-level individual dataset, i.e., the New Zealand Government’s
Integrated Data Infrastructure (IDI). The reconstructed census-based SRRS is replicable in
other countries’ population censuses and could be used to study stress-related research.
Second, in this study, we further applied this census-based SRRS instrument to explore and
test various stress-related hypotheses related to changes in residence. We hypothesised
that, ceteris paribus, (1) renters will experience less stress than owner-occupiers, regardless
of home moving; (2) a (more frequent) change in residence will (further) increase overall
stress; and (3) the change in residence will increase the stress level of owner-occupiers more
than renters.
The rest of the paper is structured as follows. Section 2provides a literature review
and develops a set of testable hypotheses. Section 3describes the construction of the
census-based SRRS and the sampling method and research design for hypothesis testing in
the natural experiment. Section 4discusses the empirical results. Section 5further discusses
the implications of the results and addresses some limitations of this research for future
study. Section 6concludes and discusses future implications.
2. Literature Review
The process of giving up an established home and social network and relocating to
another neighbourhood, city, or country is often accompanied by feelings of loss, alienation,
and fear of uncertainty. Moving for adults may be made more pleasant by the anticipation of
a more challenging or rewarding occupation or by the intellectual stimulation of relocating
to a new environment. However, most moves are not made to improve one’s life. Many
home moves are driven by significant life events such as deaths and divorces and act as an
added burden to one’s life. According to Bowlby, separation and the emotions attached
to home moves are the most difficult events with which children deal [
10
]. Disruptions
in the separation and individuation process associated with a home move could lead to
lifelong personality problems. Moving may also be a problem for the friends of a child who
is relocating. Rubin found that friends of moving children suffered increases in loneliness,
irritability, and anger following their companion’s departure [11].
Many studies of stress in humans attempt to quantify various life events as to their
stress levels. On Holmes and Rahe’s scale [
8
], changes in residence and a change in school
receive a value of 20 on a scale of 100. The death of a spouse received a full point. Life
event scores may be added to yield a total life crisis value. On this scale, a move always
adds to a person’s stress when it accompanies events such as family disintegration, loss of
job, or death. All things being equal, a move without other adversities would appear to
have a less negative impact on individuals than moves accompanied by life events such as
illness or divorce. All too often, however, a move comes about due to another life crisis.
Public health studies also have evidence that frequent moving may have health im-
plications. In a study of various housing situations, Weir et al. [12] found that individuals
who reported having two or more residences in the past six months were more likely to
engage in risky unprotected intercourse and exchanging sex for money or drugs. Likewise,
German [
13
] demonstrated that residential transience was associated with HIV drug-related
behaviours. Transient individuals also tended to be younger, have lower incomes, and
Urban Sci. 2022,6, 75 3 of 17
were less likely to have a main sexual partner than nontransient respondents, suggesting
that these individuals may have differing needs [14].
In urban science, “residential mobility” refers to interregional, intra-urban, or internal
migration [
15
]. Residential mobility is often considered as the adjustment of housing
needs for employment, family, or access to amenities [
16
]. Coulter and van Ham [
17
]
defined residential mobility through the life course framework as “a mechanism that
enables households to adjust their housing, neighbourhood and locational consumption
to meet their changing needs and preferences” (p. 1037). Understanding residential
mobility is challenging because many studies draw from aggregate census metrics, assume
only a “typical” mover and reveal only broad trends [
17
]. Whilst such studies provide
policymakers with crucial information, they have clear limitations for understanding the
complexity of movements and the well-being of diverse individuals, especially their stress
levels. In order to overcome such inadequacies in regional studies, the “new mobilities
paradigm” calls for incorporating a life-course framework [
18
]. Experiences of place and
home related to residential mobility are particular to the mobile individual and should not
be assumed but explored as lived experiences. Subsidised homeownership is, therefore,
perceived as a factor limiting residential mobility [19,20].
Life-course studies represent a research paradigm rooted in behavioural psychology
and sociology over 60 years ago [
21
]. Since then, many different disciplines, ranging
from sociology to developmental psychology, have examined the longitudinal experiences
of people in their “biography, history and of the problems of their intersection within
their social structure” [
21
] (p. 149). There are two predominant approaches in life-course
studies: one focuses on analysing how life events experienced at certain ages influence life
course, and the other on developmental outcomes resulting from the life-course change.
One major challenge in conducting life-course analysis is that the research often requires
longitudinal studies, which are survey-based, and it is challenging to obtain identical times
of measurement for all ages in the cohort [
22
]. Moreover, surveys often tend to be subjective
responses to recent life experiences, which are always contingent on participants’ age and
the time in their lives [
22
]. The accuracy of survey results is also subject to individual
stress in the extreme [
23
] and to physiological desensitisation arising from chronic allostatic
stresses [24].
Vulnerability–Stress Models: Moving Residence as a Stressful Life Event
In psychology, there are many definitions of stress. One important concept of stress
relates to significant life events that a person perceives as undesirable [
25
–
28
]. In this study,
stress is viewed as a result of life events that disrupt the stability of individual physiology,
emotion, and cognition. Moving home is one such stressor. Selye’s classic description of
stress emphasises that such events strain the adaptive capability of people and cause an
interruption of individual routines or habits [29].
Even though stress is usually conceptualised as the occurrence of “externally” ordained
processes, two sets of factors suggest the critical role of “internal” forces in the occurrence
of stress. The first set of factors is about the coping style of people. While some stressful
events could befall people, many studies have empirically demonstrated that stress often
results from individual characteristics. For instance, a person with social skill deficits (e.g.,
inappropriate critics of others) may engender fractious relationships with family, coworkers,
and acquaintances, resulting in significant stress. Therefore, vulnerable individuals may
create their own stresses [
30
]. The second set of factors is about the self-perception of
stress [
28
], i.e., stress is dependent on the individual’s appraisal of life events. Even though
many life events are perceived as stressful (e.g., the death of a loved one), individual
differences may determine the degree of stress experienced. Thus, some individuals who
perceive events as stressful may not experience as much stress as others. Indeed, many
other factors including perception [31] can affect the determination and degree of stress.
To the best of our knowledge, the empirical evidence of the impact of moving home on
tenure type (i.e., owners and renters) and moving frequency is limited. This study attempts
Urban Sci. 2022,6, 75 4 of 17
to fill this research gap by examining (1) the stress level of renters versus owner-occupiers,
regardless of home moving; (2) the effect of frequency of changing residence on overall
stress; and (3), more importantly, whether the change in residence increases the stress level
of owner-occupiers more than renters. We first theorised the relationship between moving
home and stress level by using Zubin and Spring’s vulnerability–stress model [
32
] and
then developed the whole-of-population-based SRRS index to measure the stress level of
individuals. Finally, we conducted various statistical tests to demonstrate the stress level
differences among home renters versus owner-occupiers.
3. Development of Hypotheses
Zubin and Spring [
32
], for instance, argued that “we regard [vulnerability] as a rel-
atively permanent, enduring trait” (p. 109). They continue, “The one feature that all
schizophrenics have is the ever-presence of their vulnerability” (p. 122). Intuitively put,
the idea is that people become mentally ill when the stress they face becomes more than
they can cope with. Moreover, the capability of people to deal with stress, hence their
vulnerability, varies, so problems that one person may take in their stride may cause an-
other person to be depressed or even psychotic. Figure 1demonstrates Zubin and Spring’s
vulnerability–stress model [
32
] and helps to explain the testable hypotheses formulated in
our empirical tests.
Urban Sci. 2022, 6, x FOR PEER REVIEW 4 of 17
Indeed, many other factors including perception [31] can affect the determination and de-
gree of stress.
To the best of our knowledge, the empirical evidence of the impact of moving home
on tenure type (i.e., owners and renters) and moving frequency is limited. This study at-
tempts to fill this research gap by examining (1) the stress level of renters versus owner-
occupiers, regardless of home moving; (2) the effect of frequency of changing residence
on overall stress; and (3), more importantly, whether the change in residence increases the
stress level of owner-occupiers more than renters. We first theorised the relationship be-
tween moving home and stress level by using Zubin and Spring’s vulnerability–stress
model [32] and then developed the whole-of-population-based SRRS index to measure
the stress level of individuals. Finally, we conducted various statistical tests to demon-
strate the stress level differences among home renters versus owner-occupiers.
3. Development of Hypotheses
Zubin and Spring [32], for instance, argued that “we regard [vulnerability] as a rela-
tively permanent, enduring trait” (p. 109). They continue, “The one feature that all schiz-
ophrenics have is the ever-presence of their vulnerability” (p. 122). Intuitively put, the
idea is that people become mentally ill when the stress they face becomes more than they
can cope with. Moreover, the capability of people to deal with stress, hence their vulner-
ability, varies, so problems that one person may take in their stride may cause another
person to be depressed or even psychotic. Figure 1 demonstrates Zubin and Spring’s vul-
nerability–stress model [32] and helps to explain the testable hypotheses formulated in
our empirical tests.
Figure 1. Adpated from Zubin and Spring [32]’s vulnerability–stress model in moving home.
People less vulnerable to stress need to experience a great deal of stress before be-
coming distressed and mentally ill. In contrast, people who are highly vulnerable to stress
require only a small amount of stress to “tip them over the edge” into mental illness. Stress
leads first to anxiety, which can tip the person into full-blown psychosis. As the model
states, a person with low vulnerability becomes mentally ill only at a very high-stress in-
tensity (i.e., more resilient at point R0). A vulnerable person would easily become ill de-
spite a very low-stress level (i.e., at point H’). With the same vulnerability level, a person
at a higher resilience level can tolerate more stress than a person at a lower resilience level
without becoming ill (R0 > H0 and R’ > H’ with the same vulnerability level).
In general, renters can have more flexibility regarding their residential locational
choices, while homeowners often have a misallocation problem [33]. In addition, renters
are not burdened with mortgage repayment (s). Of course, renters could be burdened by
Figure 1. Adpated from Zubin and Spring [32]’s vulnerability–stress model in moving home.
People less vulnerable to stress need to experience a great deal of stress before be-
coming distressed and mentally ill. In contrast, people who are highly vulnerable to stress
require only a small amount of stress to “tip them over the edge” into mental illness. Stress
leads first to anxiety, which can tip the person into full-blown psychosis. As the model
states, a person with low vulnerability becomes mentally ill only at a very high-stress
intensity (i.e., more resilient at point R
0
). A vulnerable person would easily become ill
despite a very low-stress level (i.e., at point H’). With the same vulnerability level, a person
at a higher resilience level can tolerate more stress than a person at a lower resilience level
without becoming ill (R0> H0and R’ > H’ with the same vulnerability level).
In general, renters can have more flexibility regarding their residential locational
choices, while homeowners often have a misallocation problem [
33
]. In addition, renters
are not burdened with mortgage repayment (s). Of course, renters could be burdened
by landlords. However, the flexibility to move for renters provides them with the room
to manoeuvre and adjust their living places whenever necessary. Homeowners are also
subject to the “asset-specificity” in their own homes, including the fixtures in their owned
Urban Sci. 2022,6, 75 5 of 17
premises. [
34
]. Less stress related to residential locational choices is expected for renters.
Thus, we hypothesised that:
Hypothesis 1 (H1) Ceteris paribus, renters are, on average, less stressed than owner-occupiers.
There is a caveat, though: the implication could only be generalisable to those countries
with a lower rate of tenant eviction. For example, the rate of initiated tenants in the United
States is the highest among the OECD countries [
35
]. This may add another layer of
complexity in analysing the stress of home moving. However, in general, H1 is applicable
to the majority of OECD countries. As OECD tenant eviction report [
35
] suggested that there
is no European country with an eviction order rate above 1% (to total tenant households),
even though Austria, France and the United Kingdom (England) were fairly close to this
level. New Zealand records a physical eviction rate of less than 0.1% of tenant households,
where tenant’s rights are protected under a comprehensive Residential Tenancy Act. That
is why we applied New Zealand as a case to study the stress of moving, which can exclude
the anxiety associated with a tenant eviction.
In addition, this study further expands the implications of Zubin and Spring’s [
32
]
vulnerability–stress model in the context of moving home. Moving home is regarded by
the general public as one of the most significant life events that cause stress on people’s vul-
nerability and are typically conceptualised as a factor that can create a mentally unhealthy
state [
7
]. While the earliest psychopathology models emphasise that vulnerability factors
are primarily genetic or biological, the concept starts to include psychological factors [
28
].
Ingram et al. [
30
] noted several core features of vulnerability that help establish a working
definition of the construct.
Thus, what causes the differences in the vulnerability of people? What makes one
person more vulnerable than others? Evidence from family studies, particularly studies
involving twins, shows that vulnerability is strongly related to their genetic makeup.
“Relatives of people with schizophrenia have a greater risk of developing the illness, the
risk being progressively higher among those more genetically similar to the person with
schizophrenia.” [
36
]. However, this is not the whole story. How a person deals with stress
and their options are often related to their environment. Anything from the state of a
person’s home to their neighbourhood can make a difference. “If people who have had
mental health problems live in a calm and relaxed home atmosphere, their problems are
less likely to return.” [
36
]. People with lots of supportive friends tend to perform better in
times of crisis than people with fewer or perhaps no other people to turn to. Social support
has long been recognised as a routine way of meeting psychological needs and enhancing
the quality of life.
One pathway through which moving home may impact people’s mental well-being
is its disruption in social relationships. Frequent residential relocation may challenge
one’s ability to maintain social networks. Specifically, individuals who move frequently
may experience less social and friend support [
37
]. Additionally, individuals who move
around may have greater difficulty developing social ties with others and easy to induce
social isolation. Under stressful circumstances, less support from social networks may
further contribute to depressive symptoms and may make it more challenging to deal
with associated stressors. Therefore, a change in residence often represents a loss of
neighbourhood/social support in coping with stress. Such change can be shown by the
inward shift of the resilience curve in Figure 1. Thus, we hypothesised that:
Hypothesis 2 (H2)
Ceteris paribus, a more frequent change in residence will increase the overall
stress level of individuals.
Enz et al. [
38
] further theorised that a life involving a major transition, such as home
moving, should give rise to a higher density of memories because the transitions give
individual events a novel backdrop. If home movers are owner-occupiers, their autobio-
graphical memories would be expected to be stronger in a specific location. The association
Urban Sci. 2022,6, 75 6 of 17
between home moves and stressors is that moving around may interfere with accessing
health and social services and resources. Such access to resources is more substantial for
homeowners. Duchon et al. [
39
] suggest that residential instability is associated with a
lack of health care and found that mobility may prevent individuals from accessing care
providers. Likewise, moving home likely disrupts consistent and appropriate mental health
support and utilisation of related social services, such as public and employment assistance.
In addition, owner-occupier movers may also interrupt informal sources of care such as
self-help and church groups. “People whose social situation protects them from stress, or
the effects of stress, fare much better than those whose social situations are stressful or that
do not protect them from the effects of stress.” [36]. Thus, we hypothesised that:
Hypothesis 3 (H3)
Ceteris paribus, the change in residence will increase the stress level of owner-
occupiers more than renters.
4. Research Design
The research design was covered in two stages. First, the SRRS within the New
Zealand Integrated Data Infrastructure (IDI) was adapted. Second, the empirical test was
conducted to investigate those developed hypotheses.
4.1. Social Readjustment Rating Scale (SRRS)
The best-known scale for comparison of stress caused by different life events is the
SRRS, a ratio measurement developed in 1963–1967 by psychiatrists Thomas Holmes and
Richard Rahe. The SRRS is collected as a self-reported survey to determine how stressed
people are in 43 significant life stress events. The life events were selected from an initial
survey in 1963 based on a questionnaire of patients at 6 Seattle hospitals determining their
demography and their recollection of major life stress events within their last 10 years.
The subsequent survey determined the average weighting of each of the life stresses, or
Life Change units (LCU), on a scale of hundred being the maximum, thus establishing the
rating scale.
In order to use the rating scale, the user sums the LCU from the checklist of major life
events to quantify the stress experienced through these events over a quarter, ranging from
a hundred points for a spouse’s death to eleven points for minor law violations [
8
]. On the
list of stressors within the SRRS, the stress of change in residence (20) or incurring a large
mortgage (37) is minor compared with the significant causes of stress such as the death of a
spouse (100), divorce (73), and marital separation (65). While the stress indicators do not
put moving to a new house near the top of the list, combined stresses often accompany a
change in residence, a phenomenon that has not been examined but can be captured by
the SRRS. Appendix Acontains the original Holmes and Rahe SRRS identified life events
and their corresponding stress scores. The table also contains whether each life event was
adapted into the IDI and which proxy criteria were used to describe the life event.
The SRRS was devised as a predictor that connects stressful life events with rates
of depression and poor health outcomes; it is shown to be applicable in modern times
and is still a tool used today [
40
]. It has been frequently challenged as a well-established
instrument and has known methodological, context, and sensitivity limitations, particularly
its cost to implement scaling up the survey [
41
]. While Hough et al. [
41
] (p. 81) raised much
criticism of SRRS, the authors indeed emphasised that “the critique of the development
and use of the SRRS has not been presented for the purpose of inhibiting the use of the
instrument in further research. The use of such a ratio measurement scale may make
the social scientist’s explanation of variance in illness much more complete, and even
in its present form, the scale has helped to accomplish that task. We would therefore
like to promote its improvement and usage.” This study aimed to introduce a scaleable
method to adopt SRRS to improve social scientists’ stress measures. Additionally, it was
confirmed to be robust enough to represent most major life stresses and as a good predictor
for future health outcomes, even across different ethnic groups [
42
–
44
]. Moreover, it was
Urban Sci. 2022,6, 75 7 of 17
found that the rank ordering of the life stresses remained consistent for both healthy adults
(
r = 0.96–0.89
) and patients (r = 0.91–0.70). Gerst et al. [
45
] indicating the SRRS has good
reliability and robustness.
The added advantage of the SRRS is that it captures stresses between family members
and household members, effectively capturing intergenerational trauma and generating
a holistic and quantifiable picture of people’s stress and social circumstances. Although
SRRS is a valuable tool for quantifying stress at individual levels, its implementation
costs are a shortcoming. The SRRS is a survey instrument that is very costly to scale up,
especially for population-level studies with a large enough sample required for sufficient
statistical power.
By using the microdata of people and households in the IDI, we reconstructed the
SRRS in census data, which can involve the whole population in New Zealand at any point
in time. Using the census-based SRRS can multiply the statistical power without incurring
the exorbitant cost of conducting an extensive longitudinal survey. The microdata allows
us to track an individual’s residence and corresponding housing tenures, i.e., movers
versus non-movers and owner-occupiers versus renters. This study indexed the SRRS of
individuals’ life events from their first home move (normalised as zero) and tested the
pre-and post-stress level of moving homes.
For further time series analysis, the study further reconciled the scores for each quarter
and/or conducted a difference-in-differences analysis across moving homeowners and
renters (treatment group) and non-movers (control group).
We constructed a difference-in-differences model to examine how the moving home
event affects the individual stress level in the IDI Datalab. Specifically, we used two
comparison groups (i.e., renters versus owners). We estimated the population difference-in-
differences, which measures the treatment effect of interest (i.e., moving home in our case).
E(Si,t|Rent,post move −E(Si,t|Rent,pre move)−
E(Si,t|owner,post move −E(Si,t|owner,pre move) = β(1)
where Sis the stress level, iis an individual renter/owner, and tis the time period. This
is the key outcome of the difference-in-differences method. We eliminated the common
trend between the groups, and the permanent differences between the groups, leaving a
straightforward estimate of the treatment effect, β[46].
4.2. New Zealand Integrated Data Infrastructure (IDI)
As previously mentioned, stress is personal and hard to investigate at an aggregate
level. Therefore, the New Zealand Integrated Data Infrastructure (IDI), based on the
micro-level individual census data, is a unique dataset for studying stress-related issues
at an individual level. The IDI contains person-centred microdata from more than thirty
governmental agencies, census surveys and non-government organisations, which is used
to support scientifically based policy research. Many significant life changes, such as health,
marital status and job changes, are all captured within the IDI, which can be translated into
the SRRS. This person-centred microdata allows us to apply the life-course framework to
study residential mobility [
18
]. In adapting SRRS into IDI, we identified 19 adult stressors
and 25 non-adult stresses/stressors as proxies for individual stress levels. Even though
some measures are not adaptable in the IDI, most of the major life stresses are captured by
this census-based SRRS. This study tested several hypotheses using the IDI-based SRRS to
measure the averaged stress levels of individuals between the ages of 19 and 54 residing in
the Auckland Region between 2013 and 2018. We assessed (H1) the overall stress level of
non-mover individuals amongst homeowners and renters, (H2) the frequency of moves
that affected the stress levels of individuals and (H3) the comparison between the stress
level before and after their home moves.
Urban Sci. 2022,6, 75 8 of 17
4.3. Sample Selection and Method of Analysis
The method of sample selection is a crucial piece of the research design. As the
IDI contains a vast amount of data from various data sources and in multiple formats
(stocks format vs. flow formats), a good understanding of the IDI’s data properties and a
representative/accurate sampling method is key to conducting robust analysis. Within the
IDI, there are variations, for each data source, in availability and accuracy over time; for
example, police data are only available from 2014 onwards, while rental housing counts via
the rental Bond database are only 60–70% accurate. Hence the data-model construction
and sampling method must account for the data source limitations.
In order to properly represent relevant mobility, a whole-of-population data model is
required to capture the people who move into and out of an area during any period of time.
If the data model does not include whole populations, then gaps appear in the data. The
construction of the movement data model involves internal government administrative
records, StatsNZ’s address notification data model, and DIA travel records. Logically, the
sample selection ought to be grounded on the most certain datasets, and the most grounded
of datasets is Census data, which captures people at their location of residence and indicates
their housing tenure.
However, accuracy is always in question. In an audit performed in relation to Tamaki
Regeneration Company over their housing stock, it was found that the social housing
tenants were under-reported in both the Census and the housing tenure; only around 55%
of the social housing had been accounted for in the Census 2018. The causes were issues
due to misreporting and non-responses characteristically prevalent in low-socio-economic
households. Regardless, census data are still regarded as the gold standard. Therefore, our
methodology used the 2013 and 2018 censuses as the primary indicator for location and
housing tenure, with a combination of other data sources, such as housing stock data and
Rental Bond data, as secondary/supplementary indicators.
There is also the issue of the homogeneity of the sample. Suppose a segment of
sampled residents stayed longer in a particular place compared with another group who
stayed for a shorter period. In that case, they are more likely to have fewer stressors before
they move for the index event and are more likely to move. Therefore, this segment has a
particular bias in the sample. Moreover, economic situations change at different periods
and regions, affecting the population’s stress. Thus, movements sampled must start from a
specific period in time, from a particular region. Additionally, the index movement period
bracket must be long enough to capture a good sample, where quarterly movement may
not be sufficient; hence a year-long aggregation period was used.
This is the basis for structuring the data for interventional analysis, utilising the first
move in 2015 as the index event. Therefore, the selected samples are individuals who
initially moved into Greater Auckland in 2013, as confirmed by the 2013 census. The
identified individuals are followed longitudinally to the 2018 census to determine whether
they have moved or not. The index measurement period of movement is within the year
2015, which means that the residents moved approximately after two years from the initial
location to any address within New Zealand. The movement and tenure capture use the
StatsNZ address notification data model linked to DIA travel data, the Housing Register,
and Housing Bond data.
Other considerations are the age of the sampled individuals: the SRRS has different
types of stress for adults and children; therefore, combining their stress scores in the
analysis would not be accurate. Additionally, as children under the age of 18 typically do
not have the agency to choose where they may move to and hence, react to socioeconomic
pressures the same way, their stress scores would not be accurate to compare. The elderly
have different life stresses, particularly health-related stresses, and many people beyond 54
have retired and moved into residential care. Therefore, the sample only selects those who
are 19 to 54 years old.
Another factor that can be controlled includes locations with a large number of resi-
dents registered against a single address. In an audit of these addresses, it has been found
Urban Sci. 2022,6, 75 9 of 17
that three main types of domicile exist. The first type is those with 80% of its residents being
older than 65; these are identified as aged care facilities. The second type is the majority
of its adult residents receiving WINZ benefits, which are identified as homeless shelters,
transitional housing, or organisations that use their addresses to help the homeless receive
WINZ payments. The third type is rental properties, which have large numbers of adult
residents with income, which are private community lodges. The residents of these types
of residences are excluded from the analysis.
The interventional analysis is conducted at two levels, the macro level and the micro
level. The data extract collects stresses over time at an individual’s event level, though it is
then aggregated differently for the two levels of analysis. At the macro level, the individual
stress scores are aggregated into quarters and averaged for the sample population segment
to create a stress profile. This is to allow time series analysis to be completed between
different population cohorts. This is different from the original SRRS methodology, which
uses aggregated yearly stresses to predict health outcomes, whereas our approach is to
determine the change in stresses before and after the individual’s change in residence. The
quarter in which the home move took place was considered 0 in the Index, while in seven
quarters before (
−
ve) and after (+ve), the move event was then compared, which is the
minimum number of aggregated points to show a trend. At the micro level, the individuals’
stress scores for each of the population segments for the seven quarters before and after the
movement event were then tested for differences in means before and after to determine
if there was a significant difference in mean; by using the macro level analysis, we can
determine the extent of the difference.
5. Empirical Results
While past studies utilise qualitative approaches to allow participants to articulate
personal experiences of place and home, they can miss how residential mobility is enacted.
Much quantitative life-course research in residential mobility remains aggregate and is
unlikely to focus on individual and personal stressors. This research adopted an established
clinical survey tool leveraging existing big-data infrastructure to quantitatively explore an
individual’s life stresses and examine the resulting behaviours and outcomes of moving
one’s home. By utilising this approach, several potential sources of stress in each move
can be inferred; this enables more visibility of lived experiences through the life course
framework perspective, perhaps facilitating actions for building resilience in different
socioeconomic groups. Limitations of the method were discussed, and further development
was suggested. As an initial baseline test, the analysis of whether the non-mover renters
experience less stress than owner-occupiers was presented.
In Figure 2, the average stress levels of non-movers renters, homeowners, and social
housing residents aged between 19 and 54 residing in urban Auckland between 2013
and 2018 are presented. The figure indicates that the mean stress level of homeowners is
significantly higher than renters (owner occupied (n= 11,499), rental (n= 2166), 2-sided
t-test p-value = 0.002). The mean stress level is slightly higher for homeowners than for
renters who remain in the same property over five years (i.e., non-movers). The figure
also shows that stress levels have a cyclical component that decreases over time when
individuals do not move. The result supports Hypothesis 1 that, on average, ceteris paribus,
renters are less stressed than owner-occupiers. It is worth noting that social housing
tenants (as shown by the grey line) have a much higher baseline stress level than both
homeowners and renters. The higher stress level among social housing tenants also tallies
with a general understanding that the stress level, such as pain, worry, sadness, and anger,
is significantly higher among low-income cohorts than among wealthy ones [
47
]. The
pattern suggests that the whole-of-population-based SRRS in this study is sensitive to
capturing the socioeconomic differences amongst various demographic groups and could
be a potential measure of economic deprivation.
Urban Sci. 2022,6, 75 10 of 17
Urban Sci. 2022, 6, x FOR PEER REVIEW 10 of 17
suggests that the whole-of-population-based SRRS in this study is sensitive to capturing
the socioeconomic differences amongst various demographic groups and could be a po-
tential measure of economic deprivation.
Figure 2. Average stress of non-movers by tenure, in time.
Additionally, it is unsurprising that there is a jump in stress at the time of the move,
which is reflected in our whole-of-population-based SRRS. Figure 3 shows the averaged
stress levels of mover-individuals (homeowners and renters only) between the ages of 19
and 54 residing in urban Auckland between 2013 and 2018, the first move after two years
of stay and by frequency of moves within five years. The jump is higher than the stress
score assigned to the change in residence (20 points), which indicates other stressors were
present at the time of the move. However, what is pertinent is that the average baseline
stress after the move event increases with the number/frequency of moves over the 5-year
period (i.e., 3-moves > 2-moves > 1-moves). This supports our hypothesis 2 that ceteris
paribus, a more frequent change in residence, will increase the overall stress level of indi-
viduals. Furthermore, the average stress after a moving event is higher than before.
Figure 3. Average stress of both owners and renters, by frequency of moves within 5 years. Notes:
(1 Move (n = 3609), 2 Move (n = 3753), 3 move (n = 1773)).
To further test our hypothesis 3, i.e., the change in residence will increase the stress
level of owner-occupiers more than renters, we compared groups of renters versus owners
before and after their home move. Figure 4 shows the average stress levels of 1-move in-
dividuals (Homeowners and renters only) between the ages of 19 and 54 years old resid-
ing in urban Auckland between 2013 and 2018, moving once after two years. The purpose
Figure 2. Average stress of non-movers by tenure, in time.
Additionally, it is unsurprising that there is a jump in stress at the time of the move,
which is reflected in our whole-of-population-based SRRS. Figure 3shows the averaged
stress levels of mover-individuals (homeowners and renters only) between the ages of 19
and 54 residing in urban Auckland between 2013 and 2018, the first move after two years of
stay and by frequency of moves within five years. The jump is higher than the stress score
assigned to the change in residence (20 points), which indicates other stressors were present
at the time of the move. However, what is pertinent is that the average baseline stress after
the move event increases with the number/frequency of moves over the 5-year period
(i.e., 3-moves > 2-moves > 1-moves). This supports our hypothesis 2 that ceteris paribus,
a more frequent change in residence, will increase the overall stress level of individuals.
Furthermore, the average stress after a moving event is higher than before.
Urban Sci. 2022, 6, x FOR PEER REVIEW 10 of 17
suggests that the whole-of-population-based SRRS in this study is sensitive to capturing
the socioeconomic differences amongst various demographic groups and could be a po-
tential measure of economic deprivation.
Figure 2. Average stress of non-movers by tenure, in time.
Additionally, it is unsurprising that there is a jump in stress at the time of the move,
which is reflected in our whole-of-population-based SRRS. Figure 3 shows the averaged
stress levels of mover-individuals (homeowners and renters only) between the ages of 19
and 54 residing in urban Auckland between 2013 and 2018, the first move after two years
of stay and by frequency of moves within five years. The jump is higher than the stress
score assigned to the change in residence (20 points), which indicates other stressors were
present at the time of the move. However, what is pertinent is that the average baseline
stress after the move event increases with the number/frequency of moves over the 5-year
period (i.e., 3-moves > 2-moves > 1-moves). This supports our hypothesis 2 that ceteris
paribus, a more frequent change in residence, will increase the overall stress level of indi-
viduals. Furthermore, the average stress after a moving event is higher than before.
Figure 3. Average stress of both owners and renters, by frequency of moves within 5 years. Notes:
(1 Move (n = 3609), 2 Move (n = 3753), 3 move (n = 1773)).
To further test our hypothesis 3, i.e., the change in residence will increase the stress
level of owner-occupiers more than renters, we compared groups of renters versus owners
before and after their home move. Figure 4 shows the average stress levels of 1-move in-
dividuals (Homeowners and renters only) between the ages of 19 and 54 years old resid-
ing in urban Auckland between 2013 and 2018, moving once after two years. The purpose
Figure 3.
Average stress of both owners and renters, by frequency of moves within 5 years. Notes:
(1 Move (n= 3609), 2 Move (n= 3753), 3 move (n= 1773)).
To further test our hypothesis 3, i.e., the change in residence will increase the stress
level of owner-occupiers more than renters, we compared groups of renters versus owners
before and after their home move. Figure 4shows the average stress levels of 1-move
individuals (Homeowners and renters only) between the ages of 19 and 54 years old
residing in urban Auckland between 2013 and 2018, moving once after two years. The
purpose was to determine whether there are differences in the stress of an individual
moving from various tenures and its time-series stress profile.
Urban Sci. 2022,6, 75 11 of 17
Urban Sci. 2022, 6, x FOR PEER REVIEW 11 of 17
was to determine whether there are differences in the stress of an individual moving from
various tenures and its time-series stress profile.
Figure 4. Average stress of one-off movers by tenure, 7-quarter before and after a move. Notes: 2-
sided t-tests are conducted amongst different housing tenures before and after a home move. The
mean stress level for an owner-to-owner increases from 22.96 to 24.91 (+1.95; p-value = 0.02303; with
n = 1770). The mean stress level for renter-to-renter only shows an insignificant increase from 22.35
to 23.49 (+1.14; p-value = 0.3332; with n = 489). For renters who become homeowners after the home
move, the stress level exhibited a significant increase in average stress level from 21.18 to 23.12
(+1.94; p-value = 0.03292; with n = 849). For the owner who becomes a renter after the move, the
stress level slightly decreases from 24 to 23.92, but the change is statistically insignificant (−0.08; p-
value = 0.4566; with n = 501).
By using Student’s t-tests at the individual level sample, we compare their stress level
before and after the home move and see if such difference is statistically significant. The
result shows that individuals who moved from owned-to-owned properties have their
stress level increase from 22.96 to 24.91 (+1.95; p-value = 0.02303); which is higher than the
insignificant increase for rent-to-rent properties 22.35 to 23.49 (+1.14; p-value = 0.3332).
This confirms our Hypothesis 3 (H3). Furthermore, one can note that the stress level of
new homeowners, who change from rent to own after the home move, also exhibited a
significant increase in average stress level from 21.18 to 23.12 (+1.94; p-value = 0.03292).
6. Discussion, Limitations, and Future Studies
The empirical results confirmed that moving homes is stressful. Our use of the whole-
of-population-based SRRS was able to assemble time-series data and generate observa-
tions that are methodologically challenging to be captured using the conventional house-
hold, income and well-being surveys, or at least with much lower cost.
In Figure 2, it is observed that there is a cyclical component to the stress being rec-
orded, visible in both the long-term renters’ and the owners’ segments. This has never
been noted in other research articles and warrants further investigation to determine
whether this is an instrument’s artefact or an actual socioeconomic phenomenon. Further-
more, when long-term renters in Figure 2 are compared with renters who moved only
once, in Figure 4, it is shown that the average baseline stress levels of those who moved
were both changing and fluctuating before the move event. This would suggest that there
are stresses/stressors creating pressure to move, perhaps from the need for home owner-
ship or housing insecurity.
This idea of the stresses/stressors on individuals to move is also reinforced in Figure
3, where it is noted that the baseline stress preceding the movement event. On average,
those who moved more frequently were higher than those who moved only once. From
Figure 4.
Average stress of one-off movers by tenure, 7-quarter before and after a move. Notes:
2-sided t-tests are conducted amongst different housing tenures before and after a home move. The
mean stress level for an owner-to-owner increases from 22.96 to 24.91 (+1.95; p-value = 0.02303; with
n= 1770). The mean stress level for renter-to-renter only shows an insignificant increase from 22.35 to
23.49 (+1.14; p-value = 0.3332; with n= 489). For renters who become homeowners after the home
move, the stress level exhibited a significant increase in average stress level from 21.18 to 23.12 (+1.94;
p-value = 0.03292; with n= 849). For the owner who becomes a renter after the move, the stress level
slightly decreases from 24 to 23.92, but the change is statistically insignificant (
−
0.08; p-value = 0.4566;
with n= 501).
By using Student’s t-tests at the individual level sample, we compare their stress level
before and after the home move and see if such difference is statistically significant. The
result shows that individuals who moved from owned-to-owned properties have their
stress level increase from 22.96 to 24.91 (+1.95; p-value = 0.02303); which is higher than
the insignificant increase for rent-to-rent properties 22.35 to 23.49 (+1.14; p-value = 0.3332).
This confirms our Hypothesis 3 (H3). Furthermore, one can note that the stress level of
new homeowners, who change from rent to own after the home move, also exhibited a
significant increase in average stress level from 21.18 to 23.12 (+1.94; p-value = 0.03292).
6. Discussion, Limitations, and Future Studies
The empirical results confirmed that moving homes is stressful. Our use of the whole-
of-population-based SRRS was able to assemble time-series data and generate observations
that are methodologically challenging to be captured using the conventional household,
income and well-being surveys, or at least with much lower cost.
In Figure 2, it is observed that there is a cyclical component to the stress being recorded,
visible in both the long-term renters’ and the owners’ segments. This has never been noted
in other research articles and warrants further investigation to determine whether this
is an instrument’s artefact or an actual socioeconomic phenomenon. Furthermore, when
long-term renters in Figure 2are compared with renters who moved only once, in Figure 4,
it is shown that the average baseline stress levels of those who moved were both changing
and fluctuating before the move event. This would suggest that there are stresses/stressors
creating pressure to move, perhaps from the need for home ownership or housing insecurity.
This idea of the stresses/stressors on individuals to move is also reinforced in Figure 3,
where it is noted that the baseline stress preceding the movement event. On average,
those who moved more frequently were higher than those who moved only once. From
the observations above, the data suggest that individuals under high-stress levels, which
may be related to housing insecurity, are predisposed to housing movement. While acute
stresses seem to result in one-off movements, chronic stresses result in more frequent
movement. This also requires further investigation.
Urban Sci. 2022,6, 75 12 of 17
Given that this dataset allows being down into the different stresses before and after
the move event, other observations can also be made. Table 1demonstrates the capability
of this SRRS instrument in analysing aspects of an individual’s circumstances/stresses,
which are traditionally not captured in conventional surveys. Another capability of this
instrument is the ability to dissect households’ socioeconomic segmentation; occupancy
of houses, incomes vs. market rent, ages of inhabitants, income, family structure, and
also changes in tenure. For example, in Figure 2, stresses experienced by social housing
individuals can also be analysed.
Table 1.
For 1-move individuals, moving between their owned property to another owned property,
the percentage changes in stress types, 7 quarters before to after the move.
Alcohol and Drug Use 433%
Abortion 350%
Retirement 57%
Victim of Violent Crimes 54%
Sexual Dysfunction 37%
Acute Hospitalisation 34%
Depression 26%
Arranged Hospitalisation 22%
Birth of Child 20%
Pregnancy 13%
Spouse Starting/Stopping work 13%
Change in Health of Family Member 6%
>70% Reduction in earned Income 3%
Divorce 2%
Change in Financial State 0%
Congenital Deformities 0%
Physical Deformities/Disability −8%
Marital Separation −12%
Change in Residences −13%
Marriage −28%
Death of a Close Family Member −50%
For owner-occupiers moving to another owned property, there seems to be an increase
in the stresses, which are not captured in traditional stress analysis, such as drug use,
childbirth and abortions. The potential explanation for this pattern is that drug use is a
mechanism to alleviate stress [
48
], and people may need to use it to reduce their stress
during home moves. Other stressors such as childbirth could be closely associated with
the motivations for home moves [
49
]. For instance, home movers want to improve the
educational outcome for existing or future children, given that homeowners are more
engaging parents [
50
]. The link between home ownership and wanting to have children
seems to be associated with stress experienced by owner-occupier movers. The marriage-
induced homeownership is also shown as a driver of housing market booms [51].
The original SRRS was useful in identifying suicide victims and attempters [
52
]. This
may be an instrument that can also be used to look at other similar social contagions,
including many socioeconomic and housing movement phenomena. Whilst not within
the scope of this paper, in Figure 2, it is noted that adult social housing occupants, who
are high in the deprivation index, experience notably more elevated stress levels, which
is not unexpected. One possible use of this instrument could be as an early warning tool
for detecting socioeconomic stress associated with high levels of deprivation. This could
be valuable to policymakers and public/private support services, who currently can only
use lagging measures that can quantify their effects 10–20 years post-events. This tool
could also potentially be used to measure the impact of changes in the socioeconomic
environment or policies on a quarterly basis. Housing market is often analysed in a so-
ciopolitical context in which the state government plays an important role [
53
]. Identifying
various stresses/stressors can provide a valuable understanding of what circumstances
Urban Sci. 2022,6, 75 13 of 17
the population experiences and how we may collectively assist. These are some future
research directions.
Even though the adapted population-based SRRS instrument and methodology in
this research seem to be sensitive enough to capture and quantify stresses to be used for
mobility research, there are limitations of use and further work which will improve this
approach. While the adapted SRRS instrument was able to capture major individual and
intergenerational life stresses/stressors, it is unable to capture the full spectrum of the
stresses/stressors included in the original SRRS, particularly stresses/stressors coming from
social and workplace sources. Since the 1960s, new stresses and psychological conditions
have been recognised and should also be included in the SRRS. Since this limitation is
acknowledged, many catchalls were built into the dataset, such as depression and violent
crime victimisation. These catchall characteristics and stress scores are estimated, untested,
and will require further testing and development.
7. Conclusions
By using a life-course perspective, combined with a well-established stress instrument,
leveraging off a big-data infrastructure, this paper demonstrates the possibility of arriving
at a new method to conceptualise the human life journey. In this study, we advanced
our understanding of the stresses of moving homes; the influence of mobility on place
experience; and the circumstances, advantages and challenges of moving home over a
resident’s lifetime. The new method resulted in an instrument that can empirically measure
the socioeconomic impact on an individual in any population segment far more quickly than
current measures and far more cheaply than conventional surveys, with better sensitivity
and ability to identify the influences on the individual.
While research has shown that home moving is detrimental to mental health, our
studies further suggest that frequent relocation and the housing tenure types, especially
owner-occupier, is a substantial contributors to stress. Therefore, the findings indicate that
housing strategies should be implemented to ensure that housing can be sustained over
time. This may include assistance programs that make housing more attainable for those
encountering mental illness. In addition, economic programs that aid individuals at risk
of losing their homes are needed. In addition to providing stable housing, mental health
services must be available, easily accessible among urban residents, and designed to remain
amenable under transient circumstances.
Engaging with new conceptualisations of home and place can help urban researchers
resolve longstanding tensions, ambiguities, and uncertainties about when, where, and why
mobility and immobility are desirable practices. The findings also potentially offer practical
levers for groups such as policymakers, urban planners, mental health professionals, and
the rank-and-file public on how to shape habits and policies and measure their immediate
impact on whole populations to minimise stress and maximise human potential. This
whole-of-population-based stress measurement approach may become an essential tool in
our collective quest to understand and ultimately achieve urban well-being.
Author Contributions:
Conceptualisation, K.-S.C. and D.W.; methodology, K.-S.C. and D.W.; soft-
ware, D.W.; validation, K.-S.C. and D.W.; formal analysis, K.-S.C. and D.W.; investigation, K.-S.C. and
D.W.; resources, K.-S.C. and D.W.; data curation, D.W.; writing—original draft preparation, K.-S.C.
and D.W.; writing—review and editing, K.-S.C.; visualisation, D.W.; supervision, K.-S.C.; project
administration, K.-S.C. All authors have read and agreed to the published version of the manuscript.
Funding:
This research project was financially supported by the Ministry of Business, Innovation
and Employment, New Zealand National Science Challenge: Building Better Homes Towns and
Cities (BBHTC).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Urban Sci. 2022,6, 75 14 of 17
Data Availability Statement:
Data may be obtained from a third party and are not publicly available.
The data used in this study are held with the Integrated Data Infrastructure and are managed by
Statistics New Zealand. These data are publicly available, although access is restricted. Please see
https://www.stats.govt.nz/integrated-data/integrated- data-infrastructure/ (accessed on 20 October
2022) for more details.
Acknowledgments:
We would like to thank Paul Rouse for connecting both authors to make this
research possible. We would also like to thank the staff of Tamaki Regeneration Company for enabling
us to conduct this research.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
Phases of
Life Life Event Stress
Score IDI Proxy Criteria
All Depression -
Not officially in SRRS but an overall catchall for stress. Uses MoH
Drugs Prescription for antidepressants
All Victim of Violent Crimes - Not officially in the SRRS, but PTS is now a recognised
stress/stressor. Sourced from Police victimisation data.
Adult Death of a Spouse 100
Spouse as defined by a current DIA Marriage/Civil Union record,
against the DIA Death records
Adult Divorce 73 Divorce as recorded in the DIA Marriage/Civil Union Records
Adult Marital separation 65
Marriage and civil unions as recorded in the DIA Marriage/Civil
Union Records
Adult Imprisonment 63 Corrections records for any incarceration event
Adult Death of a close family member 63
Family member (limited by immediate family line) as defined by
DIA Birth Records, against DIA Death records
Adult Personal injury or illness 53 MoH Acute and Arranged Inpatient hospital admissions
Adult Marriage 50 Marriage and Civil Unions as recorded in the DIA
Marriage/Civil Union Records
Adult Dismissal from work 47 IRD data, 70% drop in Wage and Salary income of the current
month is compared with the average of the last 3 months
Adult Retirement 45 Those individuals who do not have a wage or salary income for
more than a year, over 45 year old.
Adult Change in health of family member 44
Family member (limited by immediate family line) as defined by
DIA Birth Records against MoH Inpatient admission
Adult Pregnancy 40 MoH NNPAC data of first Obstetrics clinic visit OR DIA Birth
Records without a prior NNPAC record for 9 months.
Adult Sexual difficulties 39
Male: MoH Prescription of Viagra, Female: MoH Coded
diagnosis of Gynaecological disorders (endometriosis, ovarian
cysts, uterine disorders, hysterectomies)
Adult Gain a new family member 39
Family member (limited by immediate family line) as defined by
DIA Birth Records, against DIA birth records
Adult Change in financial states 38 IRD Data, 70% drop/increase in Total income of the current
month in compared with the average of the last 3 months
Adult Child leaving home 29
Adult Spouse starts or stops work 26
Spouses at current DIA marriage/civil union record, against IRD
70% drop/increase in spouses income (vs. past 3 months)
Adult Begin or end school 26 MoE School records
Adult Change in residence 20 Address changes as identified within the address notification
system
Adult Minor Viloation of law 11 Police offense data
Non Adult Death of parent 100 Death of a defined parent as recorded in the DIA Births and
Marriages records.
Non Adult Unplanned pregnancy/abortion 100
Abortion records based on clinical coding within MoH inpatient
records, OR visits to an obstetrics clinic within the MoH NNPAC
Data, before the age of 18
Urban Sci. 2022,6, 75 15 of 17
Phases of
Life Life Event Stress
Score IDI Proxy Criteria
Non Adult Getting married 95 Marriage and Civil Unions as recorded in the DIA
Marriage/Civil Union Records
Non Adult Divorce of parents 90 Divorce as recorded in the DIA Marriage/Civil Union Records,
against parental records in DIA Births and Marriages
Non Adult Acquiring a visible deformity 80 Physical Deformity codes defined by Clinical Coding within
MoH Inpatient records
Non Adult Fathering a child 70 Parental records as defined in DIA Birth records
Non Adult Jail sentence of parent for over a year 70 Corrections records for any incarceration event >1 year.
Non Adult Marital separation of parents 69 Divorce as recorded in the DIA Marriage/Civil Union Records,
for parents as defined in DIA Births and Marriages records
Non Adult Death of a brother or sister 68 Siblings as defined by connection with Parent in the DIA births,
against DIA Death records
Non Adult Unplanned pregnancy of sister 64 Siblings as defined by connection with Parent in the DIA births
and Death records, against, DIA births data.
Non Adult Marriage of parent to stepparent 63 Parental records in DIA Birth records, against DIA marriage
record not as parents
Non Adult
Having a visible congenital deformity
62 Congenital Deformity codes defined by Clinical Coding within
MoH Inpatient records
Non Adult Serious illness requiring
hospitalisation 58 MoH Acute and Arranged Inpatient hospital admissions
Non Adult Hospitalisation of parent 55 Parental records as defined in DIA Births and Marriages, against
MoH Inpatient Hospital admission records
Non Adult Jail sentence of parent for over
30 days 53 Corrections records for any incarceration event >30 days, not
including those over 1 year
Non Adult Suspension from school 50 MoE Truancy and Suspension Data
Non Adult Becoming involved with
drugs/alcohol 50 MoH PrimHD Drug and Alcohol rehabilitation Data
Non Adult Birth of a brother or sister 50 Siblings as defined by connection with Parent in the DIA births,
against DIA Birth records
Non Adult Loss of job by parent 46 IRD data, 70% drop in Wage and Salary income of the current
month in comparison with the average of the last 3 months
Non Adult Change in parent’s financial status 45 IRD Data, 70% drop/increase in Total income of the current
month in comparison with the average of the last 3 months.
Non Adult Being a senior in high school 42 MoE School data
Non Adult Hospitalisation of a sibling 41 Siblings as defined by connection with Parent in the DIA births,
against MoH Inpatient Admission data
Non Adult Brother or sister leaving home 37 Siblings as defined by connection with Parent in the DIA births,
against address notification data
Non Adult Addition of third adult to family 34
Adult as defined by individuals who are over 18yo and reside at
the same location as the individual, of the address
notification data
Non Adult Mother or father beginning work 26
IRD data, 70% increase in Wage and Salary income of the current
month in comparison with the average of the last 3 months
References
1.
Winstanley, A.; Thorns, D.C.; Perkins, H.C. Moving House, Creating Home: Exploring Residential Mobility. Hous. Stud.
2002
,17,
813–832. [CrossRef]
2.
Boston Medical Center. Housing Instability Negatively Affects the Health of Children and Caregivers: New Research Finds One
in Three Low-Income Renters Face Housing Instability at Greater Risk of Poor Health and Other Hardships. ScienceDaily. 22
January 2018. Available online: www.sciencedaily.com/releases/2018/01/180122110815.htm (accessed on 29 November 2021).
3.
Webb, R.T.; Pedersen, C.B.; Mok, P.L.H. Adverse Outcomes to Early Middle Age Linked With Childhood Residential Mobility.
Am. J. Prev. Med. 2016,51, 291–300. [CrossRef] [PubMed]
4.
Rumbold, A.R.; Giles, L.C.; Whitrow, M.J.; Steele, E.J.; E Davies, C.; Davies, M.J.; Moore, V.M. The effects of house moves during
early childhood on child mental health at age 9 years. BMC Public Heal. 2012,12, 583. [CrossRef] [PubMed]
5.
Kemper, P.; Blumenthal, D.; Corrigan, J.M.; Cunningham, P.J.; Felt, S.M.; Grossman, J.M.; Kohn, L.T.; Metcalf, E.C.; Peter, R.F.S.;
Strouse, R.C.; et al. The design of the community tracking study: A longitudinal study of health system change and its effects on
people. Inquiry 1996,33, 195–206.
Urban Sci. 2022,6, 75 16 of 17
6.
Crosswell, A.D.; Lockwood, K.G. Best practices for stress measurement: How to measure psychological stress in health research.
Health Psychol. Open 2020,7, 2055102920933072. [CrossRef]
7.
Morris, T.; Manley, D.; Northstone, K.; Sabel, C.E. How do moving and other major life events impact mental health? A
longitudinal analysis of UK children. Health Place 2017,46, 257–266. [CrossRef]
8. Holmes, T.H.; Rahe, R.H. The social readjustment rating scale. J. Psychosom. Res. 1967,11, 213–218. [CrossRef]
9.
Gibb, S.; Bycroft, C.; Matheson-Dunning, N. Identifying the New Zealand Resident Population in the Integrated Data Infrastructure
(IDI); Statistics New Zealand: Wellington, New Zealand, 2016.
10. Bowlby, J. Processes of mourning. Int. J. Psycho Anal. 1961,42, 317–340.
11. Rubin, Z. Children’s Kiendships; Harvard University Press: Cambridge, MA, USA, 1980.
12.
Weir, B.; Bard, R.S.; O’Brien, K.; Casciato, C.J.; Stark, M.J. Uncovering Patterns of HIV Risk Through Multiple Housing Measures.
AIDS Behav. 2007,11 (Suppl. 2), 31–44. [CrossRef]
13.
German, D.; Davey, M.A.; Latkin, C.A. Residential Transience and HIV Risk Behaviors Among Injection Drug Users. AIDS Behav.
2007,11 (Suppl. 2), 21–30. [CrossRef]
14.
Dong, M.; Anda, R.F.; Felitti, V.J.; Williamson, D.F.; Dube, S.R.; Brown, D.W.; Giles, W.H. Childhood residential mobility and
multiple health risks during adolescence and adulthood: The hidden role of adverse childhood experiences. Arch. Pediatr. Adolesc.
Med. 2005,159, 1104–1110. [CrossRef] [PubMed]
15.
Rossi, P.H. Why Families Move: A Study in the Social Psychology of Urban Residential Mobility; The Free Press: Glencoe, IL, USA, 1955.
16. Holdsworth, C. Family and Intimate Mobilities; Springer: Berlin/Heidelberg, Germany, 2013.
17.
Coulter, R.; Van Ham, M.; Findlay, A.M. Re-thinking residential mobility. Prog. Human Geogr.
2015
,40, 352–374. [CrossRef]
[PubMed]
18. Bailey, A. Population geography: Lifecourse matters. Prog. Hum. Geogr. 2008,33, 407–418. [CrossRef]
19.
Cheung, K.S.; Wong, S.K. Entry and exit affordability of shared equity homeownership: An international comparison. Int. J. Hous.
Mark. Anal. 2019,13, 737–752. [CrossRef]
20.
Wong, S.K.; Deng, K.K.; Cheung, K.S. Housing wealth effects for private and subsidized homeowners. Int. J. Hous. Mark. Anal.
2018,11, 771–787. [CrossRef]
21. Mills, C.W. The Sociological Imagination; Oxford University Press: New York, NY, USA, 1959.
22.
Elder, G.H. Age Differentiation and the Life Course. Annu. Rev. Sociol.
1975
,1, 165–190. Available online: http://www.jstor.org/
stable/2946043 (accessed on 20 October 2022). [CrossRef]
23.
Smith, R.C. Response Bias in the State-Trait Anxiety Inventory: Detecting the Exaggeration of Stress. J. Psychol.
1974
,86, 241–246.
[CrossRef]
24.
Coffman, J.A. Chronic stress, physiological adaptation and developmental programming of the neuroendocrine stress system.
Futur. Neurol. 2020,15, FNL39. [CrossRef]
25. Lazarus, R.S.; Folkman, S. Stress, Appraisal, and Coping; Springer publishing company: Berlin/Heidelberg, Germany, 1984.
26.
Luthar, S.S.; Zigler, E. Vulnerability and competence: A review of research on resilience in childhood. Am. J. Orthopsychiatry
1991
,
61, 6–22. [CrossRef]
27. Monroe, S.M.; Peterman, A.M. Life stress and psychopathology. Life Events Psychol. Funct. Theor. Methodol. Issues 1988, 31–63.
28.
Monroe, S.M.; Simons, A.D. Diathesis-stress theories in the context of life stress research: Implications for the depressive disorders.
Psychol. Bull. 1991,110, 406. [CrossRef] [PubMed]
29. Selye, H. The Stress of Life; McGraw-Hill Book Company: New York, NY, USA, 1956.
30. Ingram, R.E.; Miranda, J.; Segal, Z.V. Cognitive Vulnerability to Depression; Guilford Press: New York, NY, USA, 1998.
31.
Li, L.H.; Cheung, K.S.; Tse, W.S. Understanding the shoppers’ perception in retail shopping malls: A self-determination theory
perspective. J. Strateg. Mark. 2021, 1–16. [CrossRef]
32. Zubin, J.; Spring, B. Vulnerability: A new view of schizophrenia. J. Abnorm. Psychol. 1977,86, 103. [CrossRef]
33.
Cheung, K.S.; Wong, S.K.; Chau, K.W.; Yiu, C.Y. The Misallocation Problem of Subsidised Housing: A Lesson from Hong Kong.
Sustainability 2021,13, 1855. [CrossRef]
34. Wong, S.K.; Cheung, K.S. Renewing a Lease at a Discount or Premium? J. Real Estate Res. 2017,39, 215–234. [CrossRef]
35.
OECD. OECD Affordable Housing Database Section HC 3.3 EVICTIONS. Technical Report, Social Policy Division. 2021. Available
online: http://www.oecd.org/els/family/HC3-3-Evictions.pdf (accessed on 27 May 2021).
36.
Warner, R. How much of the burden of schizophrenia is alleviated by treatment? Br. J. Psychiatry
2003
,183, 375–376. [CrossRef]
[PubMed]
37.
Sluzki, C.E. Disruption and reconstruction of networks following migration/relocation. Fam. Syst. Med.
1992
,10, 359–363.
[CrossRef]
38.
Enz, K.F.; Pillemer, D.B.; Johnson, K.M. The relocation bump: Memories of middle adulthood are organised around residential
moves. J. Exp. Psychol. Gen. 2016,145, 935. [CrossRef]
39.
Duchon, L.M.; Weitzman, B.C.; Shinn, M. The Relationship of Residential Instability to Medical Care Utilization Among Poor
Mothers in New York City. Med Care 1999,37, 1282–1293. [CrossRef]
40.
Kullmann, P.; Alana, J.; Hayes, S.; Pamphlett, R. Is psychological stress a predisposing factor for amyotrophic lateral sclerosis
(ALS)? An online international case-control study of premorbid life events, occupational stress, resilience and anxiety. PLoS ONE
2018,13, e0204424.
Urban Sci. 2022,6, 75 17 of 17
41.
Hough, R.L.; Fairbank, D.T.; García, A.M. Problems in the Ratio Measurement of Life Stress. J. Health Soc. Behav.
1976
,17, 70–82.
[CrossRef] [PubMed]
42.
Hobson, C.J.; Kamen, J.; Szostek, J.; Nethercut, C.M.; Tiedmann, J.W.; Wojnarowicz, S. Stressful Life Events: A Revision and
Update of the Social Readjustment Rating Scale. Int. J. Stress Manag. 1998,5, 1–23. [CrossRef]
43.
Alejandra, D.; García, G. The Social Readjustment Rating Scale of Holmes and Rahe in Mexico: A rescaling after 16 years. Rev.
Mex. Psicol. 2012,29, 16–32.
44.
Woon, T.-H.; Wagner, N.N.; Holmes, T.H.; Masuda, M. The Social Readjustment Rating Scale: A Cross-Cultural Study of
Malaysians and Americans. J. Cross Cult. Psychol. 1971,2, 373–386. [CrossRef]
45.
Gerst, M.S.; Grant, I.; Yager, J.; Sweetwood, H. The reliability of the social readjustment rating scale: Moderate and long-term
stability. J. Psychosom. Res. 1978,22, 519–523. [CrossRef]
46. Yiu, C.Y.; Cheung, K.S. The Age–Period–Cohort Problem in Hedonic House Prices Models. Econometrics 2022,10, 4. [CrossRef]
47.
Graham, C. The High Costs of Being Poor in America: Stress, Pain, and Worry. In Brookings Institution Social Mobility Memo;
Immerwahr, J., Tony, F., Eds.; Brookings Institution: Washington, DC, USA, 2000. Available online: http://www.brookings.edu/
blogs/socialmobility-memos/posts/2015/02/19-cost-poverty-stress-graham (accessed on 20 October 2022).
48. Sinha, R. Chronic Stress, Drug Use, and Vulnerability to Addiction. Ann. New York Acad. Sci. 2008,1141, 105–130. [CrossRef]
49.
Tan, T.H.; Khong, K.W. The Link between Homeownership Motivation and Housing Satisfaction. Int. J. Econ. Manag.
2012
,
6, 1–20.
50.
Grinstein-Weiss, M.; Shanks, T.R.W.; Manturuk, K.R.; Key, C.C.; Paik, J.-G.; Greeson, J.K. Homeownership and parenting practices:
Evidence from the community advantage panel. Child. Youth Serv. Rev. 2010,32, 774–782. [CrossRef]
51.
Cheung, W.K.S.; Chan, J.T.K.; Monkkonen, P. Marriage-induced homeownership as a driver of housing booms: Evidence from
Hong Kong. Hous. Stud. 2020,35, 720–742. [CrossRef]
52.
Blasco-Fontecilla, H.; Delgado-Gomez, D.; Legido-Gil, T.; de Leon, J.; Perez-Rodriguez, M.M.; Baca-Garcia, E. Can the Holmes-
Rahe Social Readjustment Rating Scale (SRRS) Be Used as a Suicide Risk Scale? An Exploratory Study. Arch. Suicide Res.
2012
,
16, 13–28. [CrossRef] [PubMed]
53.
Li, L.H.; Cheung, K.S. Housing price and transaction intensity correlation in Hong Kong: Implications for government housing
policy. Neth. J. Hous. Built Environ. 2016,32, 269–287. [CrossRef]
Available via license: CC BY 4.0
Content may be subject to copyright.
