Heterogeneous residential preferences among millennials and members of generation X in California: A latent-class approach

Lee, Circella, Mokhtarian, and Guhathakurta, 2019, in Transportation Research Part D

doi:10.1016/j.trd.2019.08.001

Location	California
Population	Other (specify)
Sample size	729
Factor analysis type	exploratory factor analysis, Oblique rotation
Stepwise regression	no
Removal of insignificant variables	no
Reviewed by	MWC

Abstract

The millennial generation, the cohort born from 1981 to 1996, lives in large cities or denser parts of metropolitan areas more than preceding generations did at the same age. Studies have theorized that a combination of temporary economic hardship, long-term societal changes, and changing preferences and attitudes have been responsible for Millennials’ unique residential choices. This study examines a less-explored question about the presence and significance of heterogeneity in residential preferences across and within generations. In doing so, this study employs a latent-class choice model on a commuter subsample of Millennials and members of Generation X (n = 729) of the California Millennials Dataset, which collected a rich set of variables on various dimensions in Fall 2015. Using randomly-generated unlabeled choice sets at the US Census block group level, this study identifies three latent classes. The Younger, Pro-Urban Class (53% of our dataset; 66% of its millennial cases and 42% of its Gen Xers) behaves as the stereotypical Millennials in popular media, preferring urban amenities; the Affluent, Highly-Educated Class (32% of our dataset; 25% of its millennials and 38% of its Gen Xers) appears to pursue lifestyles and high socioeconomic status over homeownership or good school districts; and the Middle-Class Homeowner Class (15% of our dataset; 8% of its millennial cases and 21% of its Gen Xers) presents more traditional family-oriented suburban lifestyles. After the examination of shares of the three classes by age and neighborhood type, we provide suggestions for future research and effective planning responses. © 2019 Elsevier Ltd

Factors

Variable	Pattern loading
The functionality of a car is more important to me than its brand (Other)	0.579
To me, a car is just a way to get from place to place (Mode-liking)	0.48

Variable	Pattern loading
Greenhouse gases from human activities are creating major problems (Environmentalism)	0.796
Any climate change that may be occurring is part of a natural cycle (Environmentalism)	-0.656
It is pointless for me to try too hard to be more environmentally friendly because I am just one person (Environmentalism)	-0.307

Variable	Pattern loading
I'm already well-established in my field of work (Lifestyle)	0.704
I'm still trying to figure out my career (e.g. what I want to do, where I'll end up) (Lifestyle)	-0.636
I am generally satisfied with my life (Lifestyle)	0.387

Variable	Pattern loading
I picture myself living long-term in a suburban setting (Res preference/perception)	0.819
A house in the suburbs in the best place for kids to grow up (Res preference/perception)	0.568
I picture myself living long-term in an urban setting (Res preference/perception)	-0.31

Variable	Pattern loading
I would/do enjoy having a lot of luxury things (Lifestyle)	0.441
I prefer to minimize the material goods I possess (Lifestyle)	-0.412
For me, a lot of the fun of having something nice is showing it off (Lifestyle)	0.387
I like to be among the first people to have the latest technology (Lifestyle)	0.38
To me, owning a car is a symbol of success (Social norms)	0.316

Variable	Pattern loading
It's best to finish one project before starting another (Lifestyle)	0.518
I like to juggle two or more activities at the same time (Lifestyle)	-0.346

Variable	Pattern loading
I definitely want to own a car (Other)	0.697
I am fine with not owning a car, as long as I can use or rent one any time I need it (Convenience)	-0.5

Variable	Pattern loading
My commute is stressful (Comfort)	-0.802
My commute is generally pleasant (Mode-liking)	0.689
Traffic congestion is a major problem for me personally (Convenience)	-0.544
The time I spend commuting is generally wasted time (Convenience)	-0.501
Getting stuck in traffic does not bother me that much (Convenience)	0.305

Variable	Pattern loading
We should raise the price of gasoline to reduce the negative impacts on the environment (Environmentalism)	0.937
We should raise the price of gasoline to provide funding for better public transportation (Transport policy)	0.841
The government should put restrictions on car travel in order to reduce congestion (Transport policy)	0.331

Variable	Pattern loading
The importance of exercise is overrated (Other)	-0.822
Getting regular exercise is very important to me (Lifestyle)	0.587

Variable	Pattern loading
Social media (e.g. Facebook) makes my life more interesting (Lifestyle)	0.505
People are generally trustworthy (Other)	0.442
I enjoy the social aspects of shopping in stores (Social norms)	0.323

Variable	Pattern loading
I prefer to shop in a store rather than online (Lifestyle)	0.998
I enjoy shopping online (Lifestyle)	-0.413

Variable	Pattern loading
I prefer to live in a spacious home, even if it is farther from public transportation and many places I go to (Res preference/perception)	0.764
I prefer to live close to transit even if it means I will have a smaller home and live in a more crowded area (Res preference/perception)	-0.69
I like the idea of living somewhere with large yards and lots of space between homes (Res preference/perception)	0.428
I like the idea of having different types of businesses (such as stores, offices, restaurants, banks, and library) mixed in with the homes in my neighborhood (Res preference/perception)	-0.357

Variable	Pattern loading
The environmental impacts of the various means of transportation affect the choices I make (Environmentalism)	0.739
I am committed to using a less polluting means of transportation as much as possible (Environmentalism)	0.598
The price of fuel affects the choices I make about my daily travel (Cost)	0.532
To improve air quality, I am willing to pay a little more to use a hybrid or other clean-fuel vehicle (Environmentalism)	0.384

Variable	Pattern loading
Having Wi-Fi and/or 3G/4G connectivity everywhere I go is essential to me (Lifestyle)	0.609
Getting around is easier than ever with my smartphone (Convenience)	0.492
Learning how to use new technologies is often frustrating (Lifestyle)	-0.359
Technology creates at least as many problems as it does solutions (Lifestyle)	-0.31

Variable	Pattern loading
My schedule makes it hard or impossible for me to use public transportation (Convenience)	0.58
I am too busy to do many things I'd like to do (Convenience)	0.443
Most of the time, I have no reasonable alternative to driving (Convenience)	0.388

Variable	Pattern loading
I like trying things that are new and different (Lifestyle)	0.592
I have a strong interest in traveling to other countries (Lifestyle)	0.405

Models

Dependent variable	Census block group of residence
Model type	Latent class choice model
Sample size	729.0
R²	nan
Adjusted R²
Pseudo R² (nan)	nan
AIC	nan
BIC	nan
Log-likelihood at zero	nan
Log-likelihood at constants	nan
Log-likelihood at convergence	nan

Block group choice model: affluent, highly-educated
Variable	Coefficient	p-value
% Non-Hispanic white for a non-Hispanic white commuter	0.01	>0.1
% Non-Hispanic Asian for a non-Hispanic Asian commuter	0.08	<0.01
% Hispanic for a Hispanic commuter	-0.05	<0.01
% older Millenials, age 25-34	0.0	>0.1
Natural log of median household income, standardized	0.61	<0.01
Median home value / annual household income for owners	-0.17	<0.01
Median rent * 12 / annual household income for renters	-1.32	<0.01
Quality of elementary school (1-10)	0.04	>0.1
Natural log of distance to workplace/school	-1.53	<0.01
Amenities (factor score)	-0.07	>0.1
Land use mix (factor score)	-0.06	>0.1
Density (factor score)	-0.5	<0.01
Los Angeles region	5.16	>0.1
San Francisco Bay Area	-8.87	>0.1
Central Valley	-0.4	>0.1
San Diego region	15.06	>0.1
Sacramento region	-3.59	>0.1
Northern California and Others	-7.36	>0.1

Block group choice model: middle-class homeowners
Variable	Coefficient	p-value
% Non-Hispanic white for a non-Hispanic white commuter	-0.02	>0.1
% Non-Hispanic Asian for a non-Hispanic Asian commuter	0.17	<0.01
% Hispanic for a Hispanic commuter	0.09	<0.01
% older Millenials, age 25-34	0.01	>0.1
Natural log of median household income, standardized	1.44	<0.01
Median home value / annual household income for owners	-0.67	<0.01
Median rent * 12 / annual household income for renters	-1.14	>0.1
Quality of elementary school (1-10)	0.43	<0.01
Natural log of distance to workplace/school	-1.52	<0.01
Amenities (factor score)	-0.85	<0.01
Land use mix (factor score)	0.46	<0.01
Density (factor score)	-0.87	<0.01
Los Angeles region	4.74	>0.1
San Francisco Bay Area	-4.27	>0.1
Central Valley	-2.0	>0.1
San Diego region	5.86	>0.1
Sacramento region	-3.46	>0.1
Northern California and Others	-0.87	>0.1

Block group choice model: younger, pro-urban
Variable	Coefficient	p-value
% Non-Hispanic white for a non-Hispanic white commuter	0.02	<0.01
% Non-Hispanic Asian for a non-Hispanic Asian commuter	0.02	<0.05
% Hispanic for a Hispanic commuter	0.01	>0.1
% older Millenials, age 25-34	0.05	<0.01
Natural log of median household income, standardized	-0.01	>0.1
Median home value / annual household income for owners	-0.03	>0.1
Median rent * 12 / annual household income for renters	0.04	>0.1
Quality of elementary school (1-10)	-0.05	<0.1
Natural log of distance to workplace/school	-1.47	<0.01
Amenities (factor score)	0.14	<0.05
Land use mix (factor score)	0.31	<0.01
Density (factor score)	-0.31	<0.01
Los Angeles region	-6.03	>0.1
San Francisco Bay Area	2.83	>0.1
Central Valley	3.26	>0.1
San Diego region	-5.98	>0.1
Sacramento region	2.92	>0.1
Northern California and Others	3.0	>0.1

Class membership: affluent, highly-educated
Variable	Coefficient	p-value
Intercept	-6.14	0.004802364948378601
Non-hispanic White	2.44	0.00012814325897769763
Education: some college	-12.14	2.034166485032074e-06
Education: Bachelor's degree	4.73	2.7895445453784973e-05
Education: graduate school	5.5	1.0337061914444945e-05
Number of household children under 6	-1.22	0.10739785629623944
Number of household children 6 to 11	10.15	1.9359295919674224e-06
Number of household children 12 to 17	0.1	0.8965664266908777
Income $60,001–$120,000	-2.21	0.0046548004134630006
Income more than $120,000	7.25	6.482762574755441e-06
Tenure: homeowner	-3.94	9.870090124985964e-06
Cars per driver above 0, less than 1	7.98	0.00018402025494812513
Cars per driver = 1	6.73	0.0012379021807735757
Cars per driver greater than 1	4.43	0.02780689502699718
Pro-suburban	-1.34	0.005435889845402553
Car as tool	-2.55	3.324752745936799e-05
Pro-environmental policies	-2.22	1.3613753198749023e-05

Class membership: middle-class homeowners
Variable	Coefficient	p-value
Intercept	-6.04	0.000837783898900657
Non-hispanic White	-2.94	0.00013896679175973148
Education: some college	5.15	0.00016324754740537628
Education: Bachelor's degree	-1.81	0.0702957871680776
Education: graduate school	-5.82	0.00013345167405942604
Number of household children under 6	-1.53	0.04550026389635842
Number of household children 6 to 11	1.16	0.06288552596150532
Number of household children 12 to 17	1.3	0.08913092551708601
Income $60,001–$120,000	5.66	2.4430231850613993e-05
Income more than $120,000	-11.36	7.464303920956539e-06
Tenure: homeowner	6.2	1.2983912858199886e-06
Cars per driver above 0, less than 1	-0.76	0.5823193735766927
Cars per driver = 1	-4.77	0.002288413662045352
Cars per driver greater than 1	-1.75	0.23800021491040146
Pro-suburban	4.4	1.5093029583912454e-06
Car as tool	4.0	5.364591559198217e-06
Pro-environmental policies	0.61	0.08011831372763423

Class membership: younger, pro-urban
Variable	Coefficient	p-value
Intercept	12.18	1.1738575289577113e-06
Non-hispanic White	0.5	0.19705065809949573
Education: some college	6.99	8.587028939910368e-06
Education: Bachelor's degree	-2.92	0.0001446960878501713
Education: graduate school	0.32	0.6671956411909152
Number of household children under 6	2.75	0.0009001744811842904
Number of household children 6 to 11	-11.32	9.583665532275631e-07
Number of household children 12 to 17	-1.39	0.02320758304380699
Income $60,001–$120,000	-3.44	3.956591173648327e-05
Income more than $120,000	4.11	0.00015682835876718038
Tenure: homeowner	-2.26	3.788723990116871e-05
Cars per driver above 0, less than 1	-7.22	1.7079810942011164e-05
Cars per driver = 1	-1.96	0.07507596069703348
Cars per driver greater than 1	-2.68	0.030772669567850963
Pro-suburban	-3.06	2.034166485032074e-06
Car as tool	-1.45	0.0004315467985893662
Pro-environmental policies	1.61	4.131501382498293e-05