Title page for ETD etd-0707103-135356


Type of Document Master's Thesis
Author Nilufar, Fahmida
Author's Email Address fniluf1@lsu.edu
URN etd-0707103-135356
Title Assessing Sample Bias and Establishing Standardized Procedures for Weighting and Expansion of Travel Survey Data
Degree Master of Science (M.S.)
Department Civil & Environmental Engineering
Advisory Committee
Advisor Name Title
Chester Wilmot Committee Chair
Brian Wolshon Committee Member
Sherif Ishak Committee Member
Keywords
  • assessing bias
  • nonresponse bias
  • post stratification
  • iterative proportional fitting
  • coverage bias
  • weighting and expansion
Date of Defense 2003-06-18
Availability unrestricted
Abstract
Recent household travel surveys are encountering problems with non-response, non-coverage, non-reporting, and even incorrect or incomplete reporting of trips. In addition, data from survey samples are affected by the social, and economic conditions of the respondents, as well as the survey instrument used to collect the data. As a result, household travel survey data are invariably associated with some level of bias. These threats to the integrity of the data are often ignored while analyzing the sample survey data. However, considering the importance of information derived from such surveys, it is necessary that the survey must provide an accurate reflection of the population it represents. Given current conditions, the accuracy of travel survey data has become a matter of concern. The present study is an attempt to address the issue, and to develop standard procedures that can be applied to every household or person in a survey to assess, analyze, and adjust bias associated with that household or person. In developing these standards the study has reviewed a few recent household travel surveys in depth. The review has focused on the presence and extent of bias observed in those surveys, identified variables that were used for assessing bias, appraised the instruments and the methods of data collection used, and finally, observed the methods used for adjustment of biased data. The quality of past surveys is assessed by comparing the values (mean and proportion) of selected survey variables with those from some standard secondary data source (e.g. census, PUMS), as well as, by considering their variability (sampling) error. The study demonstrates the application of weighting and factoring (expansion) techniques on a sampled data by applying a simplified weighting and factoring techniques at the household, person, and trip levels.
Files
  Filename       Size       Approximate Download Time (Hours:Minutes:Seconds) 
 
 28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access 
  Nilufar_thesis.pdf 387.73 Kb 00:01:47 00:00:55 00:00:48 00:00:24 00:00:02

Browse All Available ETDs by ( Author | Department )

If you have more questions or technical problems, please Contact LSU-ETD Support.