BETYdb is used to manage and distribute agricultural and ecological data. This book provides instruction on entering data through the BETYdb web interface. The web interface provides a sequence of pages that walk through the process of entering meta-data, and then the option of entering trait and yield data through a similar web form or via upload of a text (csv) file. For entering large tables of data there is Bulk Upload Wizard. This is useful when entering more than a few dozen trait or yield data from a single source.
There are typically two categories of data that are entered:
- Results from previously published research, typically statistical summaries.
- Primary data, observations at the level of an individual replicate.
BETYdb supports both of these, because it was designed to support new research that quantitatively builds on previous research, allowing researchers to develop, test, and evaluate new hypotheses based on what is already known. For users interested in entering their own data, the protocol is much simpler than the dense prose implies. This is because people collecting new data a) have more complete information than is provided in scientific publications and b) the ratio of data to meta-data is much higher compared to extracting statistical summaries from the literature.
Therefore, sections on how to interpret and transform statistics, enter missing dates, extract data from figures and tables, and entering trait and yield data one at a time are helpful for meta-analysis but not necessary for primary data. For primary data enter the relevant metadata (sites, treatments, managements) and then upload a .csv file as described in the section ['Bulk Upload'](bulk upload.md) in the chapter ['Adding Trait and Yield Data'](Adding Traits and Yields.md).
This document provides a comprehensive protocol for entering previously published data into BETYdb for meta-analysis. In particular, the section 'Finding and Preparing Data' details how to manage the search, review, annotation, and extraction of information from previously published papers to facilitate the collaboration between scientists and data entry technicians.
This document has guided research teams through the process of extracting data from hundreds of published scientific articles. The general approach is to dividing the task of identifying which data to enter, which must be done by a scientist, from other steps that can be done by a data entry technician (typically undergraduate biology majors).
David LeBauer, Moein Azimi, David Bettinardi, Rachel Bonet, Emily Cheng, Michael Dietze, Patrick Mulrooney, Scott Rohde, Andy Tu