Analyzing Electricity Use of Low Income Weatherization Program Participants Using Propensity Score Analysis and a Hierarchical Linear Growth Model

<p>This evaluation utilized propensity score matching methods and a longitudinal hierarchical linear growth model to determine the effect of residential energy efficiency upgrade(s) on household electricity use for the low-income community over the course of a year in the City and County of Denver, Colorado. Propensity score analysis with risk set matching was performed at each month under analysis applying nearest neighbor and nearest neighbor with caliper approaches by balancing covariates across the treatment and control groups. Following the completion of propensity score analysis, the data were aggregated to form a data set that was used in a hierarchical linear growth model. A hierarchical linear growth model was used to examine mean differences in electricity use between the treatment groups after controlling for a set of covariates.</p> <p>Results indicated that electricity consumption was best predicted with the propensity score matched subsample. Conditional growth models produced a statistically nonsignificant difference in electricity use following residential energy efficiency upgrade(s) after controlling for variables such as sex, age, primary heating fuel, square footage of household, water heater fuel type, number of household members, type of household, status of home ownership, disability status, race, unworked income, and method of payment. None of the covariates were statistically significant in predicting electricity consumption for the subsample. As a final stage of analysis, another longitudinal hierarchical linear model was used with the entire data set, both matched and unmatched cases, to compare the results across the two data sets. The results for this model indicated a statistically significant effect of treatment with number of household members, type of dwelling, and unworked income serving as statistically significant predictors of electricity use. Since a subsample based on propensity score analysis was to simulate a randomized control trial, which is considered the gold standard in experimental research, and it is more difficult to obtain statistically significant results with a smaller sample, the results from this subsample take precedence over the results obtained from the entire sample.</p> <p>This evaluation contributes to the fields of energy efficiency and evaluation practice through the application of propensity score matching algorithms to monthly longitudinal data to be able to accurately isolate the effect of treatment on the outcome. The results from the propensity score-based sample highlight the need to utilize techniques, such as propensity score matching to control for confounding variables in a quasi-experimental study. This evaluation demonstrates that in the absence of these types of selection techniques, results could be biased. Finally, the results have informed the direction of future research and focus areas at the local level for analyzing energy efficiency programs for a low-income population.</p>

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Year dc:date.available

2017

Author dc:creator

• Polson, Ksenia

Contributors dc:contributor

• Antonio Olmos, Ph.D.
• Kathy Green
• Kellie Keeling

• Energy efficiency
• Program evaluation
• Propensity score matching
• Research methods
• Statistics and Probability

Statement dc:rights

• <p>Copyright is held by the author. User is responsible for all copyright compliance.</p>

Language dc:language

en

Repository record dc:identifier

https://digitalcommons.du.edu/etd/1275

OAI identifier oai:identifier

oai:digitalcommons.du.edu:etd-2275