1976Transportation Research Record Journal of the Transportation Research BoardRequires access

LOCATION, HOUSING, AUTOMOBILE OWNERSHIP, AND MODE TO WORK: A JOINT CHOICE MODEL

Steven R. Lerman

Open publisher page 190 citations

Abstract

Household location decisions are closely related to other choices of housing, automobile ownership, and mode to work. This paper describes a model that considers these long-run decisions, termed the mobility bundle, as jointly determined, thus eliminating the need for an arbitary set of assumptions about a sequence of choices. The model developed is based on disaggregate choice theory. Each potential location-housing-automobile ownership-mode to work combination is a distinct alternative, of which only one is selected by each house-hold. The basic methodology used is the multinomial logit model. A sample of skilled, single worker households working and residing in the Washington, D.C., metropolitan area in 1968 was used to estimate the model. The variables used to describe each alternative included locational attributes, housing attributes, transportation level of service to work, spatial opportunities for shopping trips, automobile ownership attributes, and the socioeconomic characteristics of the household. Even with the relatively small sample used, a wide range of behavioral effects were measured. It is concluded that models such as the one described here could replace existing model systems used to forecast residential location patterns. The increase in behavioral content such models permit would allow credible work-trip forecasts to be made as a part of land use forecasting. /Author/

About this research paper

What this paper is about

Household location decisions are closely related to other choices of housing, automobile ownership, and mode to work. This paper describes a model that considers these long-run decisions, termed the mobility bundle, as jointly determined, thus eliminating the need for an arbitary set of assumptions about a sequence of choices. The model developed is based on disaggregate choice theory. Each potential location-housing-automobile ownership-mode to work combination is a distinct alternative, of which only one is selected by each house-hold. The basic methodology used is the multinomial logit model. A sample of skilled, single worker households working and residing in the Washington, D.C., metropolitan area in 1968 was used to estimate the model. The variables used to describe each alternative included locational attributes, housing attributes, transportation level of service to work, spatial opportunities for shopping trips, automobile ownership attributes, and the socioeconomic characteristics of the household. Even with the relatively small sample used, a wide range of behavioral effects were measured. It is concluded that models such as the one described here could replace existing model systems used to forecast residential location patterns. The increase in behavioral content such models permit would allow credible work-trip forecasts to be made as a part of land use forecasting. /Author/

Why it matters

OpenAlex reports 190 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Household location decisions are closely related to other choices of housing, automobile ownership, and mode to work. This paper describes a model that considers these long-run decisions, termed the mobility bundle, as jointly determined, thus eliminating the need for an arbitary set of assumptions about a sequence of choices. The model developed is based on disaggregate choice theory. Each potential location-housing-automobile ownership-mode to work combination is a distinct alternative, of which only one is selected by each house-hold. The basic methodology used is the multinomial logit model. A sample of skilled, single worker households working and residing in the Washington, D.C., metropolitan area in 1968 was used to estimate the model. The variables used to describe each alternative included locational attributes, housing attributes, transportation level of service to work, spatial opportunities for shopping trips, automobile ownership attributes, and the socioeconomic characteristics of the household. Even with the relatively small sample used, a wide range of behavioral effects were measured. It is concluded that models such as the one described here could replace existing model systems used to forecast residential location patterns. The increase in behavioral content such models permit would allow credible work-trip forecasts to be made as a part of land use forecasting. /Author/

Key concepts: Multinomial logistic regression, Mode choice, Sample (material), Car ownership, TRIPS architecture, Metropolitan area, Work (physics), Mode (computer interface)

Related papers

Back to paper searchBrowse research topicsOriginal source
LOCATION, HOUSING, AUTOMOBILE OWNERSHIP, AND MODE TO WORK: A JOINT CHOICE MODEL — Research Paper | ScholarLens