Online Program

2018 CBECS Stratum Misclassification: Issue and Solutions (308141)

Keywords: sampling frame, stratum misclassification, commercial, buildings, CBECS

A major goal of the 2018 Commercial Buildings Energy Consumption Survey (CBECS) sampling frames construction process was to efficiently oversample buildings greater than 200,000 square feet. To meet this objective, CBECS used a multi-frame design (i.e., multiple lists of large buildings combined with a multi-stage area probability frame), as done in previous cycles. Prior CBECS cycles showed that commercial building size was correlated with key statistics of interest such as total annual energy use. As a consequence, the 2018 sample design and prior designs used a building size categorical variable to help create the relatively homogenous sampling strata for the building-level selection stage. Not unexpectedly, the building size frame variable had significant classification or measurement error. As a result, the sample design lost some efficiency and experienced a loss in precision due to variation in weights within respondent reported building size categories. This research explores the misclassification rates by comparing for all completed cases the building size category assigned on the sampling frame to the building size category reported by the respondent. We make comparisons by sampling frame in order to pinpoint which frame generated the highest misclassification rates. Based on these results, we then suggest two competing sample design modifications that may generate a more efficient design in future CBECS cycles.