Search

Search Results

• 31. [Article] Wider Dissemination of Household Travel Survey Data Using Geographical Perturbation Methods

  Public agencies spend vast amounts of money collecting information about passenger travel in household travel surveys. These data are valuable for the rich and detailed information they provide, which ...

  Citation × Citation Citation Abstract Copy Title: Wider Dissemination of Household Travel Survey Data Using Geographical Perturbation Methods Author: Clifton, Kelly J., Gehrke, Steven R. Year: 2014 Public agencies spend vast amounts of money collecting information about passenger travel in household travel surveys. These data are valuable for the rich and detailed information they provide, which contribute to regional and statewide travel demand models. These data have utility beyond travel demand modeling in their application to transportation policy and travel behavior research. As the demand on these data increase, so have the quantity of information collected. Detailed geospatial referencing of the home, work and other travel destinations are common practice and permit the integration with other spatially archived data sources, such as land use characteristics, transportation system information, and other built environment, social and economic data. Other public agencies, private consultancies, non-profits and educational institutions may benefit from access to the original data with applications to areas such public health, equity, transportation safety and urban planning. Oregon Modeling Collaborative (OMC) has entered into an agreement with Oregon Modeling Steering Committee (OMSC) to host and make available Oregon Household Activity Survey (OHAS) datasets. But wide distribution of these important and expensive data is limited by the requirement to protect the confidentiality of survey participants, who are guaranteed anonymity in exchange for participation. Data are often aggregated to a geographic level such as Census tracts or transportation analysis zones (TAZs) before disseminating to the public, which limits the utility of this information. This project aims to develop a methodology to permit dissemination of these spatially-explicit data to a wider range of public constituents while at the same time, protecting the identities of study participants. Making use of these data, this project will use geographical perturbation methods to add noise to the original data to protect confidentiality while at the same time allowing the detailed geo-spatial referencing to be included in the disseminated data. The process includes: The coordinates (locations) will be geographically perturbed (masked) so that they do not reveal the identity of the traveler while at the same time retaining spatial relevance and resolution. Here the limits of the perturbation (minimum and maximum displacement) need to be determined that a) ensure confidentiality and b) minimize the errors introduced to the data. The original and perturbed (masked) data will be compared using various statistical approaches to develop a set of confidence measures for different types of transportation applications. For example, geographically perturbed (masked) may be more robust for automobile travel measures, such as vehicle miles traveled (VMT) or transit accessibility than for detailed pedestrian or bicycle trip attributes. This exercise will allow spatially-explicit OHAS data to be released to the public with some information about the confidence to which it can be applied. These data will be archived at Portland State University and made available to the public. The information and algorithms can be shared with other agencies collecting and archiving travel data, such as the National Household Travel Survey, metropolitan planning organizations and state departments of transportation, to permit wider dissemination of their data. Title: Wider Dissemination of Household Travel Survey Data Using Geographical Perturbation Methods Author: Clifton, Kelly J., Gehrke, Steven R. Year: 2014 Public agencies spend vast amounts of money collecting information about passenger travel in household travel surveys. These data are valuable for the rich and detailed information they provide, which contribute to regional and statewide travel demand models. These data have utility beyond travel demand modeling in their application to transportation policy and travel behavior research. As the demand on these data increase, so have the quantity of information collected. Detailed geospatial referencing of the home, work and other travel destinations are common practice and permit the integration with other spatially archived data sources, such as land use characteristics, transportation system information, and other built environment, social and economic data. Other public agencies, private consultancies, non-profits and educational institutions may benefit from access to the original data with applications to areas such public health, equity, transportation safety and urban planning. Oregon Modeling Collaborative (OMC) has entered into an agreement with Oregon Modeling Steering Committee (OMSC) to host and make available Oregon Household Activity Survey (OHAS) datasets. But wide distribution of these important and expensive data is limited by the requirement to protect the confidentiality of survey participants, who are guaranteed anonymity in exchange for participation. Data are often aggregated to a geographic level such as Census tracts or transportation analysis zones (TAZs) before disseminating to the public, which limits the utility of this information. This project aims to develop a methodology to permit dissemination of these spatially-explicit data to a wider range of public constituents while at the same time, protecting the identities of study participants. Making use of these data, this project will use geographical perturbation methods to add noise to the original data to protect confidentiality while at the same time allowing the detailed geo-spatial referencing to be included in the disseminated data. The process includes: The coordinates (locations) will be geographically perturbed (masked) so that they do not reveal the identity of the traveler while at the same time retaining spatial relevance and resolution. Here the limits of the perturbation (minimum and maximum displacement) need to be determined that a) ensure confidentiality and b) minimize the errors introduced to the data. The original and perturbed (masked) data will be compared using various statistical approaches to develop a set of confidence measures for different types of transportation applications. For example, geographically perturbed (masked) may be more robust for automobile travel measures, such as vehicle miles traveled (VMT) or transit accessibility than for detailed pedestrian or bicycle trip attributes. This exercise will allow spatially-explicit OHAS data to be released to the public with some information about the confidence to which it can be applied. These data will be archived at Portland State University and made available to the public. The information and algorithms can be shared with other agencies collecting and archiving travel data, such as the National Household Travel Survey, metropolitan planning organizations and state departments of transportation, to permit wider dissemination of their data. Close

• 32. [Article] Do TODs Make a Difference?

  In this report, we present research that measures the outcomes of TOD areas in relation to their metropolitan area controls with respect to (1) jobs by sector; (2) housing choice for household types based ...

  Citation × Citation Citation Abstract Copy Title: Do TODs Make a Difference? Author: Nelson, Arthur C., Miller, Matt, Eskic, Dejan, Kim, Keuntae, Ewing, Reid, Liu, Jenny H., Berggren, Matt, Mumuni, Zakari Year: 2015 In this report, we present research that measures the outcomes of TOD areas in relation to their metropolitan area controls with respect to (1) jobs by sector; (2) housing choice for household types based on key demographic characteristics; (3) housing affordability based on transportation costs; and (4) job-worker balance as a measure of accessibility. Prior literature has not systematically evaluated TOD outcomes in these respects with respect to light rail transit (LRT), commuter rail transit (CRT), bus rapid transit (BRT), and streetcar transit (SCT) systems. Our analysis helps close some of these gaps. We apply our analysis to 23 fixed-guideway transit systems operating in 17 metropolitan areas in the South and West that have one or more of those systems. We find: (1) Most TOD areas gained jobs in the office, knowledge, education, health care and entertainment sectors, adding more than $100 billion in wages capitalized over time; (2) In assessing economic resilience associated with LRT systems, jobs continued to shift away from TOD areas before the Great Recession, the pace slowed during the recession but reversed during recovery, leading us to speculate that LRT TOD areas may have transformed metropolitan economies served by LRT systems; (3) Rents for offices, retail stores and apartments were higher when closer to SCT systems, had mixed results with respect LRT systems, but were mostly lower with respect to CRT systems (our BRT sample was too small to evaluate); (4) SCT systems performed best in terms of increasing their TOD-area shares of metropolitan population, households and householders by age, housing units, and renters with BRT systems performing less well while LRT and CRT systems experienced a much smaller shift in the share of growth; (5) Household transportation costs as a share of budgets increase with respect to distance from LRT transit stations to seven miles, suggesting the proximity to LRT stations reduces total household transportation costs; (6) Emerging trends that may favor higher-wage jobs locating in TOD areas over time than lower- or middle-wage jobs, perhaps because TOD areas attract more investment which requires more productive, higher-paid labor to justify the investment; and (7) The share of workers who commute 10 minutes or less to work increases nearly one-half of 1 percent for each half-mile their resident block group is to an LRT transit station, capping at a gain of 1.3 percent, which is not a trivial gain. Our report summarizes case studies of 23 transit systems and three journal articles based on our research. Title: Do TODs Make a Difference? Author: Nelson, Arthur C., Miller, Matt, Eskic, Dejan, Kim, Keuntae, Ewing, Reid, Liu, Jenny H., Berggren, Matt, Mumuni, Zakari Year: 2015 In this report, we present research that measures the outcomes of TOD areas in relation to their metropolitan area controls with respect to (1) jobs by sector; (2) housing choice for household types based on key demographic characteristics; (3) housing affordability based on transportation costs; and (4) job-worker balance as a measure of accessibility. Prior literature has not systematically evaluated TOD outcomes in these respects with respect to light rail transit (LRT), commuter rail transit (CRT), bus rapid transit (BRT), and streetcar transit (SCT) systems. Our analysis helps close some of these gaps. We apply our analysis to 23 fixed-guideway transit systems operating in 17 metropolitan areas in the South and West that have one or more of those systems. We find: (1) Most TOD areas gained jobs in the office, knowledge, education, health care and entertainment sectors, adding more than $100 billion in wages capitalized over time; (2) In assessing economic resilience associated with LRT systems, jobs continued to shift away from TOD areas before the Great Recession, the pace slowed during the recession but reversed during recovery, leading us to speculate that LRT TOD areas may have transformed metropolitan economies served by LRT systems; (3) Rents for offices, retail stores and apartments were higher when closer to SCT systems, had mixed results with respect LRT systems, but were mostly lower with respect to CRT systems (our BRT sample was too small to evaluate); (4) SCT systems performed best in terms of increasing their TOD-area shares of metropolitan population, households and householders by age, housing units, and renters with BRT systems performing less well while LRT and CRT systems experienced a much smaller shift in the share of growth; (5) Household transportation costs as a share of budgets increase with respect to distance from LRT transit stations to seven miles, suggesting the proximity to LRT stations reduces total household transportation costs; (6) Emerging trends that may favor higher-wage jobs locating in TOD areas over time than lower- or middle-wage jobs, perhaps because TOD areas attract more investment which requires more productive, higher-paid labor to justify the investment; and (7) The share of workers who commute 10 minutes or less to work increases nearly one-half of 1 percent for each half-mile their resident block group is to an LRT transit station, capping at a gain of 1.3 percent, which is not a trivial gain. Our report summarizes case studies of 23 transit systems and three journal articles based on our research. Close

• 33. [Article] Multiday GPS Travel Behavior Data for Travel Analysis: The Effect of Day-to-Day Travel Time Variability on Auto Travel Choices

  This project explored the potential of archived multi-day GPS data to expand the understanding of travel-time reliability. While reliability is often observed and considered at the system or segment level, ...

  Citation × Citation Citation Abstract Copy Title: Multiday GPS Travel Behavior Data for Travel Analysis: The Effect of Day-to-Day Travel Time Variability on Auto Travel Choices Author: Dill, Jennifer, Broach, Joseph Paul, Deutsch-Burgne, Kate, Xu, Yanzhi, Guensler, Randall, Levinson, David, Tang, Wenyun Year: 2014 This project explored the potential of archived multi-day GPS data to expand the understanding of travel-time reliability. While reliability is often observed and considered at the system or segment level, travel-time uncertainty is also experienced at the household and trip level. Any move toward incorporating reliability into regional travel models will necessitate a re-examination of travel-time variation at more disaggregate levels. The analysis uses multiday vehicle-based GPS data analyzed within the Transportation Secure Data Center (TSDC). There were three major goals for the research. The first goal was to consider the ways in which multiday GPS data could be translated into data on reliability. The second goal was to explore relationships between trip- and household-level travel-time reliability and related trip, household, network, and urban location factors. The third goal involved use of the topic of reliability as a case study to test the usefulness of the TSDC in its current form for academic research. Title: Multiday GPS Travel Behavior Data for Travel Analysis: The Effect of Day-to-Day Travel Time Variability on Auto Travel Choices Author: Dill, Jennifer, Broach, Joseph Paul, Deutsch-Burgne, Kate, Xu, Yanzhi, Guensler, Randall, Levinson, David, Tang, Wenyun Year: 2014 This project explored the potential of archived multi-day GPS data to expand the understanding of travel-time reliability. While reliability is often observed and considered at the system or segment level, travel-time uncertainty is also experienced at the household and trip level. Any move toward incorporating reliability into regional travel models will necessitate a re-examination of travel-time variation at more disaggregate levels. The analysis uses multiday vehicle-based GPS data analyzed within the Transportation Secure Data Center (TSDC). There were three major goals for the research. The first goal was to consider the ways in which multiday GPS data could be translated into data on reliability. The second goal was to explore relationships between trip- and household-level travel-time reliability and related trip, household, network, and urban location factors. The third goal involved use of the topic of reliability as a case study to test the usefulness of the TSDC in its current form for academic research. Close

• 34. [Article] Development of Destination Choice Models for Pedestrian Travel

  Most research on walking behavior has focused on mode choice or walk trip frequency. In contrast, this study is one of the first to analyze and model the destination choice behaviors of pedestrians within ...

  Citation × Citation Citation Abstract Copy Title: Development of Destination Choice Models for Pedestrian Travel Author: Clifton, Kelly J., Singleton, Patrick A., Muhs, Christopher D., Schneider, Robert J. Year: 2016 Most research on walking behavior has focused on mode choice or walk trip frequency. In contrast, this study is one of the first to analyze and model the destination choice behaviors of pedestrians within an entire region. Using about 4500 walk trips from a 2011 household travel survey in the Portland, Oregon, region, we estimated multinomial logit pedestrian destination choice models for six trip purposes. Independent variables included terms for impedance (walk trip distance), size (employment by type, households), supportive pedestrian environments (parks, a pedestrian index of the environment variable called PIE), barriers to walking (terrain, industrial-type employment), and traveler characteristics. Unique to this study was the use of small-scale destination zone alternatives. Distance was a significant deterrent to pedestrian destination choice, and people in carless or childless households were less sensitive to distance for some purposes. Employment (especially retail) was a strong attractor: doubling the number of jobs nearly doubled the odds of choosing a destination for home-based shopping walk trips. More attractive pedestrian environments were also positively associated with pedestrian destination choice after controlling for other factors. These results shed light on determinants of pedestrian destination choice behaviors, and sensitivities in the models highlight potential policy-levers to increase walking activity. In addition, the destination choice models can be applied in practice within existing regional travel demand models or as pedestrian planning tools to evaluate land use and transportation policy and investment scenarios. Title: Development of Destination Choice Models for Pedestrian Travel Author: Clifton, Kelly J., Singleton, Patrick A., Muhs, Christopher D., Schneider, Robert J. Year: 2016 Most research on walking behavior has focused on mode choice or walk trip frequency. In contrast, this study is one of the first to analyze and model the destination choice behaviors of pedestrians within an entire region. Using about 4500 walk trips from a 2011 household travel survey in the Portland, Oregon, region, we estimated multinomial logit pedestrian destination choice models for six trip purposes. Independent variables included terms for impedance (walk trip distance), size (employment by type, households), supportive pedestrian environments (parks, a pedestrian index of the environment variable called PIE), barriers to walking (terrain, industrial-type employment), and traveler characteristics. Unique to this study was the use of small-scale destination zone alternatives. Distance was a significant deterrent to pedestrian destination choice, and people in carless or childless households were less sensitive to distance for some purposes. Employment (especially retail) was a strong attractor: doubling the number of jobs nearly doubled the odds of choosing a destination for home-based shopping walk trips. More attractive pedestrian environments were also positively associated with pedestrian destination choice after controlling for other factors. These results shed light on determinants of pedestrian destination choice behaviors, and sensitivities in the models highlight potential policy-levers to increase walking activity. In addition, the destination choice models can be applied in practice within existing regional travel demand models or as pedestrian planning tools to evaluate land use and transportation policy and investment scenarios. Close

• 35. [Article] Development of a Pedestrian Demand Estimation Tool

  Most research on walking behavior has focused on mode choice or walk-trip frequency. In contrast, this study is one of the first to analyze the destination choice behaviors of pedestrians. Using about ...

  Citation × Citation Citation Abstract Copy Title: Development of a Pedestrian Demand Estimation Tool Author: Clifton, Kelly, Singleton, Patrick Allen, Muhs, Christopher D., Schneider, Robert J. Year: 2015 Most research on walking behavior has focused on mode choice or walk-trip frequency. In contrast, this study is one of the first to analyze the destination choice behaviors of pedestrians. Using about 4,500 walk trips from a 2011 household travel survey in the Portland, OR, region, we estimated multinomial logit pedestrian destination choice models for six trip purposes. Independent variables included terms for impedance (walk-trip distance); size (employment by type, households); supportive pedestrian environments (parks, a pedestrian index of the environment variable called PIE); barriers to walking (terrain, industrial-type employment); and traveler characteristics. Unique to this study was the use of small-scale destination zone alternatives. Distance was a significant deterrent to pedestrian destination choice, and people in carless or childless households were less sensitive to distance for some purposes. Employment (especially retail) was a strong attractor: doubling the number of jobs nearly doubled the odds of choosing a destination for home-based shopping walk trips. More attractive pedestrian environments were also positively associated with pedestrian destination choice after controlling for other factors. These results shed light on determinants of pedestrian destination choice behavior, and sensitivities in the models highlight potential policy-levers to increase walking activity. In addition, the destination choice models can be used in regional travel demand models or as pedestrian planning tools to evaluate land use and transportation policy and investment scenarios. Title: Development of a Pedestrian Demand Estimation Tool Author: Clifton, Kelly, Singleton, Patrick Allen, Muhs, Christopher D., Schneider, Robert J. Year: 2015 Most research on walking behavior has focused on mode choice or walk-trip frequency. In contrast, this study is one of the first to analyze the destination choice behaviors of pedestrians. Using about 4,500 walk trips from a 2011 household travel survey in the Portland, OR, region, we estimated multinomial logit pedestrian destination choice models for six trip purposes. Independent variables included terms for impedance (walk-trip distance); size (employment by type, households); supportive pedestrian environments (parks, a pedestrian index of the environment variable called PIE); barriers to walking (terrain, industrial-type employment); and traveler characteristics. Unique to this study was the use of small-scale destination zone alternatives. Distance was a significant deterrent to pedestrian destination choice, and people in carless or childless households were less sensitive to distance for some purposes. Employment (especially retail) was a strong attractor: doubling the number of jobs nearly doubled the odds of choosing a destination for home-based shopping walk trips. More attractive pedestrian environments were also positively associated with pedestrian destination choice after controlling for other factors. These results shed light on determinants of pedestrian destination choice behavior, and sensitivities in the models highlight potential policy-levers to increase walking activity. In addition, the destination choice models can be used in regional travel demand models or as pedestrian planning tools to evaluate land use and transportation policy and investment scenarios. Close

• 36. [Article] Combining Sensor Monitoring and Ethnography to Evaluate Household Latrine Usage in Rural India

  This paper advances research on methods used to evaluate sanitation usage and behavior. The research used quantitative and qualitative methods to contribute to new understanding of sanitation practices ...

  Citation × Citation Citation Abstract Copy Title: Combining Sensor Monitoring and Ethnography to Evaluate Household Latrine Usage in Rural India Author: O'Reilly, Kathleen, Louis, Elizabeth, Thomas, Evan A., Sinha, Antara Year: 2015 This paper advances research on methods used to evaluate sanitation usage and behavior. The research used quantitative and qualitative methods to contribute to new understanding of sanitation practices and meanings in rural India. We estimated latrine usage behavior through ethnographic interviews and sensor monitoring, specifically the latest generation of infrared toilet sensors, Portland State University Passive Latrine Use Monitors (PLUMs). Two hundred and fifty-eight rural households in West Bengal (WB) and Himachal Pradesh, India, participated in the study by allowing PLUMs to be installed in their houses for a minimum of 6 days. Six hundred interviews were taken in these households, and in others, where sensors had not been installed. Ethnographic and observational methods were used to capture the different defecation habits and their meanings in the two study sites. Those data framed the analysis of the PLUM raw data for each location. PLUMs provided reliable, quantitative verification. Interviews elicited unique information and proved essential to understanding and maximizing the PLUM data set. The combined methodological approach produced key findings that latrines in rural WB were used only for defecation, and that low cost, pit latrines were being used sustainably in both study areas. Title: Combining Sensor Monitoring and Ethnography to Evaluate Household Latrine Usage in Rural India Author: O'Reilly, Kathleen, Louis, Elizabeth, Thomas, Evan A., Sinha, Antara Year: 2015 This paper advances research on methods used to evaluate sanitation usage and behavior. The research used quantitative and qualitative methods to contribute to new understanding of sanitation practices and meanings in rural India. We estimated latrine usage behavior through ethnographic interviews and sensor monitoring, specifically the latest generation of infrared toilet sensors, Portland State University Passive Latrine Use Monitors (PLUMs). Two hundred and fifty-eight rural households in West Bengal (WB) and Himachal Pradesh, India, participated in the study by allowing PLUMs to be installed in their houses for a minimum of 6 days. Six hundred interviews were taken in these households, and in others, where sensors had not been installed. Ethnographic and observational methods were used to capture the different defecation habits and their meanings in the two study sites. Those data framed the analysis of the PLUM raw data for each location. PLUMs provided reliable, quantitative verification. Interviews elicited unique information and proved essential to understanding and maximizing the PLUM data set. The combined methodological approach produced key findings that latrines in rural WB were used only for defecation, and that low cost, pit latrines were being used sustainably in both study areas. Close

• 37. [Article] Network Indicators of the Social Ecology of Adolescents in Relative and Non-Relative Foster Households

  Though the presence, composition, and quality of social relationships—particularly as found in family networks—has an important influence on adolescent well-being, little is known about the social ecology ...

  Citation × Citation Citation Abstract Copy Title: Network Indicators of the Social Ecology of Adolescents in Relative and Non-Relative Foster Households Author: Blakeslee, Jennifer E., Kothari, Brianne H., McBeath, Bowen, Sorenson, Paul, Bank, Lew Year: 2017 Though the presence, composition, and quality of social relationships—particularly as found in family networks—has an important influence on adolescent well-being, little is known about the social ecology of youth in foster care. This study examined the social networks of foster youth participating in a large RCT of an intervention for siblings in foster care. Youth reported on the people they lived with and the relatives they were in contact with, which provided indicators of network size, composition, and relationship quality. Cluster analysis was used to identify five family network profiles for youth living in foster homes. Two identified subgroups reflected robust family networks where youth were living with relative caregiver(s) and related youth, and also reported multiple family ties outside the household, including with biological parents. The remaining three profiles reflected youth reports of fewer family connections within or beyond the foster household, with distinctions by whether they lived with siblings and/or reported having positive relationships with their mothers and/or fathers. The identified network profiles were validated using youth- and caregiver-reported measures of mental health functioning, with increased caregiver report of post-traumatic stress symptoms indicated for the three subgroups that were not characterized by a robust family network. © 2016 Elsevier Ltd Title: Network Indicators of the Social Ecology of Adolescents in Relative and Non-Relative Foster Households Author: Blakeslee, Jennifer E., Kothari, Brianne H., McBeath, Bowen, Sorenson, Paul, Bank, Lew Year: 2017 Though the presence, composition, and quality of social relationships—particularly as found in family networks—has an important influence on adolescent well-being, little is known about the social ecology of youth in foster care. This study examined the social networks of foster youth participating in a large RCT of an intervention for siblings in foster care. Youth reported on the people they lived with and the relatives they were in contact with, which provided indicators of network size, composition, and relationship quality. Cluster analysis was used to identify five family network profiles for youth living in foster homes. Two identified subgroups reflected robust family networks where youth were living with relative caregiver(s) and related youth, and also reported multiple family ties outside the household, including with biological parents. The remaining three profiles reflected youth reports of fewer family connections within or beyond the foster household, with distinctions by whether they lived with siblings and/or reported having positive relationships with their mothers and/or fathers. The identified network profiles were validated using youth- and caregiver-reported measures of mental health functioning, with increased caregiver report of post-traumatic stress symptoms indicated for the three subgroups that were not characterized by a robust family network. © 2016 Elsevier Ltd Close

• 38. [Article] Household Water Filter Use Characterization in Rural Rwanda: Signal Interpretation, Development and Validation

  Access to safe drinking water is an important health factor in many developing countries. Studies have shown that unsafe drinking water and poor sanitation practices leads to diarrheal disease, which is ...

  Citation × Citation Citation Abstract Copy Title: Household Water Filter Use Characterization in Rural Rwanda: Signal Interpretation, Development and Validation Author: Tellez Sanchez, Sarita Lucia Year: 2016 Access to safe drinking water is an important health factor in many developing countries. Studies have shown that unsafe drinking water and poor sanitation practices leads to diarrheal disease, which is one of the leading causes of death of children under five in developing countries. Provision and proper use of household water filters have been shown to effectively improve health. This thesis is focused on the refinement and validation of algorithms for data collected from pressure transducer sensors that are used in household water filters (the Vestergaard Frandsen LifeStraw Family 2.0) deployed in Rwanda by the social enterprise DelAgua Health. Statistical and signal processing techniques were used to detect the use of the LifeStraw water filters and to estimate the amount of water filtered at the time of usage. An algorithm developed by Dr. Carson Wick at Georgia Institute of Technology was the baseline for the analysis of the data. The algorithm was then refined based on data collected in the SweetLab at Portland State University, which was then applied to field data. Laboratory results indicated that the mean error of the improved algorithm is 11.5% as compared with the baseline algorithm mean error of 39%. The validation of the algorithm with field data yielded a mean error of 5%. Errors may be attributed to real-world behavior of the water filter, electronic noise, ambient temperature, and variations in the approximation made to the field data. This work also presents some consideration of the algorithm applied to soft-sided water backpacks. Title: Household Water Filter Use Characterization in Rural Rwanda: Signal Interpretation, Development and Validation Author: Tellez Sanchez, Sarita Lucia Year: 2016 Access to safe drinking water is an important health factor in many developing countries. Studies have shown that unsafe drinking water and poor sanitation practices leads to diarrheal disease, which is one of the leading causes of death of children under five in developing countries. Provision and proper use of household water filters have been shown to effectively improve health. This thesis is focused on the refinement and validation of algorithms for data collected from pressure transducer sensors that are used in household water filters (the Vestergaard Frandsen LifeStraw Family 2.0) deployed in Rwanda by the social enterprise DelAgua Health. Statistical and signal processing techniques were used to detect the use of the LifeStraw water filters and to estimate the amount of water filtered at the time of usage. An algorithm developed by Dr. Carson Wick at Georgia Institute of Technology was the baseline for the analysis of the data. The algorithm was then refined based on data collected in the SweetLab at Portland State University, which was then applied to field data. Laboratory results indicated that the mean error of the improved algorithm is 11.5% as compared with the baseline algorithm mean error of 39%. The validation of the algorithm with field data yielded a mean error of 5%. Errors may be attributed to real-world behavior of the water filter, electronic noise, ambient temperature, and variations in the approximation made to the field data. This work also presents some consideration of the algorithm applied to soft-sided water backpacks. Close

• 39. [Article] Comparison of Respondent-reported and Sensor-recorded Latrine Utilization Measures in Rural Bangladesh: A Cross-sectional Study

  Background: Health improvements realized through sanitation are likely achieved through high levels of facilities utilization by all household members. However, measurements of sanitation often rely on ...

  Citation × Citation Citation Abstract Copy Title: Comparison of Respondent-reported and Sensor-recorded Latrine Utilization Measures in Rural Bangladesh: A Cross-sectional Study Author: Delea, Maryann G., Nagel, Corey L., Thomas, Evan A., Halder, Amal K., Amin, Nuhu, Shoab, Abul K., Freeman, Matthew C., Unicomb, Leanne, Clasen, Thomas F. Year: 2017 Background: Health improvements realized through sanitation are likely achieved through high levels of facilities utilization by all household members. However, measurements of sanitation often rely on either the presence of latrines, which does not guarantee use, or respondent-reported utilization of sanitation facilities, which is prone to response bias. Overstatement of sanitation metrics limits the accuracy of program outcome measures, and has implications for the interpretation of related health impact data. Methods: We conducted a cross-sectional study of 213 households in 14 village water, sanitation and hygiene committee clusters throughout rural Bangladesh and used a combined data- and relationship-scale approach to assess agreement between respondent-reported latrine utilization and sensor-recorded measurement. Results: Four-day household-level respondent-reported defecation averaged 28 events (inter-quartile range [IQR] 20–40), while sensor-recorded defecation averaged 17 events (IQR 11–29). Comparative analyses suggest moderately high accuracy (bias correction factor=0.84), but imprecision in the data (broad scatter of data, Pearson’s r=0.35) and thus only weak concordance between measures (ρc=0.29 [95% BCa CI 0.15 to 0.43]). Conclusions: Respondent-reported latrine utilization data should be interpreted with caution, as evidence suggests use is exaggerated. Coupling reported utilization data with objective measures of use may aid in the estimation of latrine use. Title: Comparison of Respondent-reported and Sensor-recorded Latrine Utilization Measures in Rural Bangladesh: A Cross-sectional Study Author: Delea, Maryann G., Nagel, Corey L., Thomas, Evan A., Halder, Amal K., Amin, Nuhu, Shoab, Abul K., Freeman, Matthew C., Unicomb, Leanne, Clasen, Thomas F. Year: 2017 Background: Health improvements realized through sanitation are likely achieved through high levels of facilities utilization by all household members. However, measurements of sanitation often rely on either the presence of latrines, which does not guarantee use, or respondent-reported utilization of sanitation facilities, which is prone to response bias. Overstatement of sanitation metrics limits the accuracy of program outcome measures, and has implications for the interpretation of related health impact data. Methods: We conducted a cross-sectional study of 213 households in 14 village water, sanitation and hygiene committee clusters throughout rural Bangladesh and used a combined data- and relationship-scale approach to assess agreement between respondent-reported latrine utilization and sensor-recorded measurement. Results: Four-day household-level respondent-reported defecation averaged 28 events (inter-quartile range [IQR] 20–40), while sensor-recorded defecation averaged 17 events (IQR 11–29). Comparative analyses suggest moderately high accuracy (bias correction factor=0.84), but imprecision in the data (broad scatter of data, Pearson’s r=0.35) and thus only weak concordance between measures (ρc=0.29 [95% BCa CI 0.15 to 0.43]). Conclusions: Respondent-reported latrine utilization data should be interpreted with caution, as evidence suggests use is exaggerated. Coupling reported utilization data with objective measures of use may aid in the estimation of latrine use. Close

• 40. [Article] Behavioral Reactivity Associated With Electronic Monitoring of Environmental Health Interventions-A Cluster Randomized Trial with Water Filters and Cookstoves

  Subject reactivity—when research participants change their behavior in response to being observed—has been documented showing the effect of human observers. Electronics sensors are increasingly used to ...

  Citation × Citation Citation Abstract Copy Title: Behavioral Reactivity Associated With Electronic Monitoring of Environmental Health Interventions-A Cluster Randomized Trial with Water Filters and Cookstoves Author: Thomas, Evan A., Tellez-Sanchez, Sarita, Wick, Carson, Kirby, Miles Year: 2016 Subject reactivity—when research participants change their behavior in response to being observed—has been documented showing the effect of human observers. Electronics sensors are increasingly used to monitor environmental health interventions, but the effect of sensors on behavior has not been assessed. We conducted a cluster randomized controlled trial in Rwanda among 170 households (70 blinded to the presence of the sensor, 100 open) testing whether awareness of an electronic monitor would result in a difference in weekly use of household water filters and improved cookstoves over a four-week surveillance period. A 63% increase in number of uses of the water filter per week between the groups was observed in week 1, an average of 4.4 times in the open group and 2.83 times in the blind group, declining in week 4 to an insignificant 55% difference of 2.82 uses in the open, and 1.93 in the blind. There were no significant differences in the number of stove uses per week between the two groups. For both filters and stoves, use decreased in both groups over four-week installation periods. This study suggests behavioral monitoring should attempt to account for reactivity to awareness of electronic monitors that persists for weeks or more. Title: Behavioral Reactivity Associated With Electronic Monitoring of Environmental Health Interventions-A Cluster Randomized Trial with Water Filters and Cookstoves Author: Thomas, Evan A., Tellez-Sanchez, Sarita, Wick, Carson, Kirby, Miles Year: 2016 Subject reactivity—when research participants change their behavior in response to being observed—has been documented showing the effect of human observers. Electronics sensors are increasingly used to monitor environmental health interventions, but the effect of sensors on behavior has not been assessed. We conducted a cluster randomized controlled trial in Rwanda among 170 households (70 blinded to the presence of the sensor, 100 open) testing whether awareness of an electronic monitor would result in a difference in weekly use of household water filters and improved cookstoves over a four-week surveillance period. A 63% increase in number of uses of the water filter per week between the groups was observed in week 1, an average of 4.4 times in the open group and 2.83 times in the blind group, declining in week 4 to an insignificant 55% difference of 2.82 uses in the open, and 1.93 in the blind. There were no significant differences in the number of stove uses per week between the two groups. For both filters and stoves, use decreased in both groups over four-week installation periods. This study suggests behavioral monitoring should attempt to account for reactivity to awareness of electronic monitors that persists for weeks or more. Close