Fig 1.
Data types in different agricultural disciplines.
The horizontal barplot on top showing the number of respondents that assigned themselves to each of the different agricultural disciplines (Q4). The vertical barplot on the right shows the number of respondents working with the different types of data (Q10). The heatmap shows the relative distribution of data types among the different research categories (N = 191). For better comparability of the various research categories, the values were centered and scaled in each column (i.e. research categories).
Fig 2.
a) Percentage of data providers publishing their data always, sometimes or never in one of the mentioned ways (independent publication, supplementing a research paper or by providing metadata (i.e. data itself is not published). N = 144; N = 145, N = 143 respectively (Q16) b) Reasons for not publishing research among data providers that cannot make their data available are data protection, business interests or ethical reasons: regularly, occasionally, never or unknown: N = 141 (Q18).
Fig 3.
a) How often do data providers and users use different available search tools to search for relevant data and literature publications (N = 149–153; Q37)? b) Respondents’ assessment of how they rate the formulation of clear and sensitive search terms when searching for relevant data and articles, especially in interdisciplinary context (N = 160; Q38).
Fig 4.
Important data quality aspects.
Likert plot of data quality aspects that are important to data users (N = 67–126; Q29).
Table 1.
Technical components supporting publication of research data.
Table 2.
Incentives to encourage researchers to publish data more frequently at their institution.
Fig 5.
Wordcloud with data for which the respondents could imagine using a validation tool.
The size represents the frequency with which individual data types were mentioned, e.g. “measurements” was mentioned twelve times (respondents N = 66, 104 mentioned data types, 47 different data types; Q31).