-
Notifications
You must be signed in to change notification settings - Fork 0
Survey
Results from the LDA and database analysis were ground-truthed with an electronic survey sent to researchers who work throughout LAC. Survey methodology is a sound assessment of ideal, individual values despite some logistical challenges (Mostert, 2018). Survey questions were about scientists’ research experience, publication history, and opinions on funding and interdisciplinarity in the field of water resources.
Researchers received IRB exemption for protocol 1335782-1 which described the survey questions, consent waiver, and research methodology. The survey included 16 questions, was available in three languages (English, Spanish, and Portuguese), and took approximately 4 minutes to complete.
The survey was sent to corresponding authors from 22,324 papers, a subset of the final corpus, using Qualtrics survey distribution software. The breakdown of email contacts according to language of paper were 20,332 English, 1,293 Spanish, and 699 Portuguese. The survey was sent for the first time in November, 2018 and re-sent to non-respondents weekly until February 2019, when there was no longer a significant rate of response from weekly emails. Emails of inquiry were sent directly to Alyssa DeVincentis who responded to questions, comments, and concerns.
The survey response data was cleaned iteratively as new responses came in to address errors in the survey. During the first week of data collection, respondents commented that there was an 'error' in the question regarding number of publications - it did not allow respondents to answer more than 20. In response to the comments, the question was edited in Qualtrics and entries of those who commented with their actual publication number were edited. However, nearly 100 responses were collected before this error was detected and we cannot be confident in the accuracy of this data, assuming that some respondents may have published more than 20 papers but did not indicate a correction as a comment.
The survey analysis involved multiple steps. The survey responses in Spanish and Portuguese were translated to English. The document was then cleaned and imported into Atlas.ti. Responses were then coded and analyzed. Issues related to access to funding and to research collaboration were found to be particularly important to respondents.
A short summary of survey responses was compiled and sent to respondents that indicated they wanted to be contacted in the future in the initial survey. The report will serve as a thank you to the respondents and an effort to inquire if people are genuinely interested in collaborating as this project moves forward. There may be opportunity to hold webinars, workshops, or other research activities with interested survey respondents, pending future funding.