Mining temporal data for information is often inhibited by a multitude of formats: regular or irregular time intervals, point events that need aggregating, multiple observational units or repeated measurements on multiple individuals, and heterogeneous data types. This work presents a cohesive and conceptual framework for organizing and manipulating temporal data, which in turn flows into visualization, modeling, and forecasting routines. Tidy data principles are extended to temporal data by: (1) mapping the semantics of a dataset into its physical layout; (2) including an explicitly declared “index” variable representing time; (3) incorporating a “key” comprising single or multiple variables to uniquely identify units over time. This tidy data representation most naturally supports thinking of operations on the data as building blocks, forming part of a “data pipeline” in time-based contexts. A sound data pipeline facilitates a fluent workflow for analyzing temporal data. The infrastructure of tidy temporal data has been implemented in the R package, called tsibble. Supplementary materials for this article are available online.

This dataset has been formatted using Tidy data principles (Wickham 2014). That is, each observational unit is stored in its own table, each observation is a row, and each variable is a column. There is a meta-data tab included in the Excel workbook that explains all variables in each table. We have also included the R code required to import these data and perform all analyses in the manuscript.

Wickham, H. (2014). Tidy data. Journal of Statistical Software 59:10.

There are four data tables (csv documents) provided here which are formatted according to tidy data principles. We identified sites centered on known murrelet nest trees (nest sites) and on randomly selected trees (control sites) and the first table contains variables that vary at the site level. Each site had up to three stations where occupancy surveys were collected, so the second table contains variables varying at the station level. Each station was surveyed multiple times, and the third table contains variables varying at the survey level. Lastly, we had a camera trained on the nests at our active nest sites and the fourth table contains one row for every recorded action taken by an adult or nestling.

V2X Value Stream meta-study dataset accompanying the publication “Vehicle-to-Anything (V2X) Energy Services, Value Streams, and Regulatory Policy Implications” submitted to Energy Policy.

This dataset consists of a meta-study of annual economic valuations of various energy services which have been identified as being able to be provided by Vehicle-to-Anything (V2X) resources. The dataset has been created to achieve greater granularity while incorporating 15 unique studies of energy market potential while maintaining the principles of tidy data to allow for more complex visualizations and analysis.

For more complete explanations see Data Log in the Supplementary Materials of the original paper.

This dataset has been formatted using Tidy data principles (Wickham 2014). That is, each observational unit is stored in its own table, each observation is a row, and each variable is a column. There is a meta-data tab included in the Excel workbook (data) that explains all variables in each table, and Appendix C of the manuscript's supplementary materials provides additional details about how moderators were aggregated and quantified. We have also included the R code (dataAnalysis) required to import these data and perform all analyses in the manuscript. Finally, the file output.nex is a random sample of 1000 Ericson backbone trees (Jetz et al. 2012; http://birdtree.org/) that is used to calculate an average phylogenetic tree in the analyses.

Jetz, W., G. H. Thomas, J. B. Joy, K. Harmann, and A. O. Mooers (2012). The global diversity of birds in space and time. Nature 491:444-448.

Wickham, H. (2014). Tidy data. Journal of Statistical Software 59:10.

Demonstrating and modeling changes in ecosystem processes in the laboratory classroom can be logistically difficult and expensive. This complexity often leaves little time for students to generate and test hypotheses. Yet, we must foster student understanding of how matter and energy move through ecosystems to develop an appreciation of how current ecosystems function and how human-mediated global change may alter ecosystem processes. In this lesson, we describe an adaptation of the Tea Bag Index (TBI) that provides students with an inexpensive, adaptable, and easily replicated method for testing how an ecosystem function (i.e., decomposition by microorganisms) alters carbon flow between two carbon pools (i.e., dead organic matter and the atmosphere). We outline the steps that small student research groups can take to develop testable research questions with an emphasis on how abiotic factors (e.g., temperature, moisture availability) can influence the rate of biomass loss. We outline the equipment and methods that can be used for conceptual add-ons (e.g., CO2 gas analysis) and include exercises that work on teaching students principles of tidy data organizing and data analysis. Finally, we include rubrics for written and graph-based assignments and an example dataset to assist instructors in implementing the lab in their own courses. In post-lab evaluations, students reflected positively on this lab exercise in open-ended course evaluation prompts and we observed better quality data collection and analysis in subsequent experimental labs, likely motivated by the practice and guidelines provided in this lab module.

Primary Image: The biomass of dead plants as an energy source for multiple decomposers. Dead organic matter is a rich energy source for those fungi and bacteria that can break down cellulose. In this picture, multiple species of fungal decomposers work to break down a fallen log in Mt. Spokane State Park (Washington, USA). One species of Xeromphalina fungi (Division: Basidiomycota; Class: Agaricomycetes) has derived enough energy from decomposition to produce a fruiting body (dusky orange caps and stems) that this fungus will use to spread its spores. Photo Credit: Brian Connolly.

The table follows tidy data principles, with each column a variable and each row an observation. Each numeric variable shows the mean±SD. SacNumber refers to the first and second saccade after the head impulse onset. SCC indicates the semicircular canal. NaN values indicate insufficient data to determine a value. (CSV)