Uploading datasets

To create new datasets, multiple paths exist. In the web application, you can upload files to create datasets. From crunch, there are two methods for creating datasets: one for data.frames and one for files of other (non-R) formats.

The crunch package contains a sample dataset that we’ll use throughout the vignettes: a sample drawn from a political survey fielded by YouGov on behalf of the Economist.

load("../vignettes/economist.RData")
dim(economist)

## [1] 250  61

You can create a dataset from any data.frame you have in your R session with newDataset. Let’s use that sample dataset:

ds <- newDataset(economist, name="Economist/YouGov Weekly Survey")
dim(ds)

## [1] 250  61

newDataset translates R data types into their analogous types in Crunch.

• character –> Text Variable
• numeric, integer –> Numeric Variable
• factor –> Categorical Variable
• Date –> Datetime Variable

It takes a data.frame as its input; alternatively, if you have an SPSS or CSV file, you can upload it with that without first reading it into R by giving newDataset the file name. In this case, it essentially does what you would do in the web application: uploads your file and creates a dataset from it.

Loading existing Crunch datasets

Datasets already existing on the Crunch server can be loaded with loadDataset. The function takes either the dataset’s name, or the position within the dataset list returned by listDatasets:

listDatasets()

## [1] "Economist/YouGov Weekly Survey"

ds <- loadDataset("Economist/YouGov Weekly Survey")

is.dataset(ds)

## [1] TRUE

Dataset properties

Dataset have metadata beyond what a data.frame has. Datasets have a human-readable name, which you specified when you created it, and a description.

name(ds)

## [1] "Economist/YouGov Weekly Survey"

description(ds)

## [1] ""

Both can be set with <- assignment. Let’s give our dataset an informative description:

description(ds) <- "U.S. nationally representative sample, 1000 respondents"
description(ds)

## [1] "U.S. nationally representative sample, 1000 respondents"

Note that this assignment doesn’t just modify our local dataset object: it sends the new description to the server. If we pull a fresh copy of the dataset from the server, with refresh, we’ll see the description is there:

ds <- refresh(ds)
description(ds)

## [1] "U.S. nationally representative sample, 1000 respondents"

Archiving and deleting datasets

Datasets that you don’t need anymore can be either archived or deleted. Archiving removes the dataset from the primary listings of datasets, but it is not a permanently destructive action. You can archive and restore archived datasets in the web application.

Datasets can also be deleted permanently. This action cannot be undone, so it should not be done lightly. crunch provides two ways to delete a dataset: a delete method on a dataset object, like

## Not run
delete(ds)

The second way to delete is deleteDataset, which works like loadDataset: you supply a dataset name. This way is faster if you have not already loaded the dataset object into your R session: no need to fetch something from the server just to then tell the server to delete it.

For details on both, see their help pages.

Next: variable metadata