ICSD Data Components

Raw data from three of the ICSD component surveys are available for download here.

The second generation ICSD contains imputed data for all cities with populations over 50,000 (per the 2010 census) that responded to at least one of the 7 ICSD component surveys.  The imputed data is presented in a stacked structure where approximately 900 variables across 683 city observations are stacked on top of each other 20 times. 

      Variable _mj is the index variable that tells us whether data is imputed.

            _mj = 1 is the first set of imputations

            _mj = 20 is the 20th set of imputations

We have removed the raw data from the posted file to adhere to data sharing agreements made with the administrators of some of the ICSD's seven component surveys.

Analysis done with imputed data should be done with care in order to account for embedded uncertainty. Please refer to both the technical write-up describing how this data was imputed and the article Dealing with Missing Data: A Comparative Exploration of Approaches Using the Integrated City Sustainability Database for information describing how to use it.

Please cite the following if using this data: Curley, Cali, Rachel M. Krause, R.C. Feiock, Christopher V. Hawkins. (2017). Dealing with Missing Data: A comparative exploration of approaches utilizing the Integrated City Sustainability Database.  Urban Affairs Review. 55 (2): 591-615.