National Coral Reef Monitoring Program: Socioeconomic surveys of human use, knowledge, attitudes, and perceptions in the Commonwealth of the Northern Mariana Islands (CNMI) from 2016-08 to 2017-04 (NCEI Accession 0170927)

Resident surveys took place on the islands of CNMI in 2016 and 2017. CNMI contains 3 inhabited islands: Saipan, Tinian, and Rota, and these islands were the sampling strata for NCRMP in CNMI. The potential respondent universe for this study was adults, eighteen years or older, who live on Saipan, Tinian, or Rota. Due to the importance of understanding all potential users of the coral reefs who may be affected by activities related to NOAA’s CRCP, the survey was not restricted to those who live directly on the coastline. Therefore, all adults on the island were included in the potential respondent universe. Residents of CNMI over the age of 18 were surveyed via telephone and via face to face interviews from August 2016 to April 2017. To obtain a sample that did not over or under-sample regions of Saipan based on population density, the 2010 US Census population density data was initially used to develop a density dependent sampling strategy. It was quickly determined from field efforts that the 2010 population data for Saipan does not accurately represent present day population distribution. This is in part due to Typhoon Soudelor, which hit Saipan directly in August of 2015 and severely impacted infrastructure and housing. Face to face surveyors found that many of the areas in which the 2010 census data indicated high population density were still largely abandoned as of early 2017. Without accurate density information, the survey method was changed to a random approach in which the surveyors arbitrarily chose a section of the island from digital maps and drove in this area until an inhabited neighborhood was found. If more than one residence was present, the surveyor would approach every third door. If the approached residence was abandoned, the surveyor noted this and continued to the next residence in the sequence. Face to face interviewers used electronic tablets or laptops with the survey instrument programmed into an application that recorded time and duration of survey, the results of each house visited (i.e. no answer, refused survey, successful survey), and automatically skipped questions as necessary based upon answers to previous questions. Phone numbers were obtained from online phone books for Saipan, Tinian and Rota and compiled into a project database. Both landline and cell phone numbers were available. The database was then filtered by phone number prefix to remove the majority of non-residential numbers and categorize by island. Surveyors worked within a specific island until its quota had been met. The database was used to randomly select phone numbers, as well as track which numbers had been used and the results of each call (i.e. no answer, disconnected, refused survey, successful survey, non-residential number). In the case of Saipan, numbers that were initially recorded as “no answer” were called again after all available numbers had been contacted once, because more surveys were needed. Numbers which were still “no answer” were re-entered into the re-call pool and used again as necessary. The CNMI online phone book was updated during the survey process, which increased the availability of working numbers and enabled the completion of the surveys. Contracted surveyors used Computer Assisted Telephone Interviewing (CATI) software for telephone surveys and offered the survey in four languages: English, Chamorro, Carolinian, and Tagalog. A total of 722 interviews were completed (702 by phone and 20 by face to face interview), yielding a response rate of approximately 20% (20% for telephone and 29% for face to face interviews).