Reference-dependent housing choice behaviour: why are older people reluctant to move?

Figures & data

Figure 1. Print screen of a choice set.

Table 1. Attributes and attribute levels.

Attribute		Level 1	Level 2	Level 3
Distance to green^a	DCE	More than 15 minutes cycling	15 minutes cycling	10 minutes walking
	Exist	More than 15 minutes cycling	More than 10 minutes walking, but less than 15 minutes cycling	Less than 10 minutes walking
Shops^b	DCE	More than 15 minutes cycling	15 minutes cycling	10 minutes walking
	Exist	More than 15 minutes cycling	More than 10 minutes walking, but less than 15 minutes cycling	Less than 10 minutes walking
Public transport	DCE	No Bus/tram in front of the door; No train station on a 10 min car trip	Bus/tram in front of the door; No train station on a 10 min car trip	Bus/tram in front of the door; train station on a 10 min car trip
	Exist	Station in more than 15 minutes cycling	Station in more than 10 minutes walking, but less than 15 minutes cycling	Station in less than 10 minutes walking
Location	DCE	Within a larger city	Smaller city, more than 15 minutes driving from a larger city	Suburbs of a larger city
	Exist	Within a larger city	Smaller city, more than 15 minutes driving from a larger city	Suburbs of a larger city
Price	DCE	10% more expensive	Equal to the current price	10% less expensive
Neighbourhood	DCE	Mixed tenure types, mixed household types	Mostly owner-occupied, mostly seniors	Mostly owner-occupied, mixed household types

^(a) Green is defined as the nearest public green space that is suitable for walking and/or cycling (park, forest, nature reserve, etc.).

^(b) Shops is defined as the nearest location with more than 20 shops and restaurants.

Table 2. Description of the respondents.

Socio-economic characteristics of the respondent		Current dwelling and moving preferences
Aged 65 to 69	63%	Lives in an apartment	15%
Aged 70 to 74	37%	Lives in a one-family dwelling	85%
Couple	86%	Size dwelling, median	120m^2
Alone	14%
Low educated	22%	Size of garden, for one-family dwellings, median	100m^2
Middle educated	38%
High educated	40%
Household yearly gross income less than €30,000	3%	Value of the dwelling, as specified by the fiscal authorities, median	€225,000
Household yearly gross income €30,000 to €50,000	71%
Household yearly gross income more than €50,000	26%
Retired	87%

Source: Ossokina et al. (2020).

Table 3. Decrease in fit of general model when delta parameters are omitted one attribute at a time.

Attribute	LL	Decrease in fit (general vs standard) (A)			Decrease in fit (general vs constant shift) (B)
		Chisq	df	p-value	Chisq	df	p-value
General model (base)	−1181.2
Location	−1202.1	41.771	6	< 0.0001	39.534	4	< 0.0001
Green	−1185.4	8.260	6	0.220	1.254	4	0.869
Shops	−1187.4	12.265	6	0.056	8.295	4	0.081
Public transport	−1189.6	16.684	6	0.011	7.391	4	0.117

Table 4. Decrease in fit of the constant shift model when delta parameters are omitted one attribute at a time.

		Decrease in fit
Attribute	LL	Chisq	df	p-value
Constant shift model (base)	−1209.6
Location	−1210.7	2.251	2	0.325
Green	−1213.0	6.971	2	0.031
Shops	−1211.3	3.421	2	0.181
Public transport	−1214.0	8.869	2	0.012

Table 5. Estimation results.

Attribute		DCE-level	Current (reference)	Standard model		Final model
			level	Value	t-val.	Value	t-val.
Constant	α			−0.659***	−3.62	−0.364	−1.38
Location	β1	City	–	−0.300**	−2.43	−0.165	−0.77
	β2	Town (base)	–	0		0
	β3	Suburb	–	0.019	0.16	–0.431**	−2.03
	δ11	City	City			0
	δ12	City	Town			−0.725***	−3.08
	δ13	City	Suburb			−0.732***	−3.09
	δ21	Town	City			−1.054***	−4.38
	δ22	Town	Town			0
	δ23	Town	Suburb			−0.235	−1.14
	δ31	Suburb	City			−0.008	−0.04
	δ32	Suburb	Town			0.312	1.47
	δ33	Suburb	Suburb			0
Green	β1	Far	–	−0.285**	−2.31	−0.303**	−2.41
	β2	Medium (base)	–	0		0
	β3	Close	–	0.209*	1.79	0.187	1.59
	δ1	–	Far			1.108**	2.42
	δ2	–	Medium			0
	δ3	–	Close			−0.011	−0.07
Shops	β1	Far	–	−0.485***	−3.68	−0.473***	−3.54
	β2	Medium (base)	–	0		0
	β3	Close	–	0.431***	3.74	0.443***	3.79
Public transport	β1	Far	–	−0.838***	−6.12	–0.897***	−6.45
	β2	Medium (base)	–	0		0
	β3	Close	–	0.520***	4.68	0.506***	4.48
	δ1	–	Far			0.287**	2.17
	δ2	–	Medium			0
	δ3	–	Close			−0.319*	−1.71
Neighbourhood	β1	Mixed tenure & household types	–	−0.351***	−2.80	−0.374***	−2.96
	β2	Mostly home owners & mostly seniors (base)	–	0		0
	β3	Mostly home owners, mixed household types	–	−0.091	−0.76	−0.143	−1.17
Price	β1	+10%	–	−0.241*	−1.93	−0.236*	−1.87
	β2	Equal (base)	–	0		0
	β3	−10%	–	0.074	0.62	0.036	0.3
Loglikelihood				−1225.4		−1191.6
# parameters				13		23
Rho-squared				0.183		0.205
Rho-squared adjusted				0.174		0.189

*** significant at 1%, ^** significant at 5%, ^* significant at 10%.

Figure 2. Neighbourhoods of Eindhoven region (Netherlands), predicted attractivity for seniors.

Ossokina, I. V., Arentze, T. A., Van Gameren, D. & van den Heuvel, D. (2020) Best living concepts for elderly homeowners: Combining a stated choice experiment with architectural design, Journal of Housing and the Built Environment, 35, pp. 847–865.

 Web of Science ®Google Scholar