Reproducible-Research | Examples of reproducible research in linguistics

Reproducible-Research is a HTML library. Reproducible-Research has no bugs, it has no vulnerabilities and it has low support. You can download it from GitHub.

All scientific and scholarly endeavour crucially builds on holding the author accountable. To put it in financial terms: research should have an 'audit trail' of everything that lead to a particular conclusion. This is nothing new, it is basically just plain old being explicit about sources, clear about the method, and precise in citation. In more recent technical discussion, the term Reproducible Research has become en-vogue. However, there are at least three different kinds of reproducibility:. Recreation is the standard approach to reproducibility in experimental fields of research. In contrast, recreation is not possible in fields of research where data is not (easily) replicable, either because of high costs associated with its collection or because the data are historically bound (e.g., archaeological artifacts, fossils, or manuscripts). The traditional humanities scholar will normally focus on accountability. However, as more an more data is becoming available and automatic processing becomes feasible, the focus is slowly shifting towards emulation, i.e. Note that the precise prerequisites for Reproducible Research are currently heavily in flux, and it is even unclear what extend of documentation is necessary for it to be counted as "reproducible" (cf. FitzJohn et al. 2014). For some articles it might be feasible to use literature programming (Knuth 1984) by including executable code inside the text of the article (as for example implemented by the Sweave (Leisch 2002) or rMarkDown). For others, a "makefile" approach as initiated by Jon Claerbout (Schwab et al. 2000) might be more suitable. Still other articles might simply depend on explicit textual description of the steps necessary to reproduce the results as presented in the paper. A central criterium for the acceptance of papers is simply whether the reviewers are able to reproduce the results based on the documentation provided.