This function imports one dataset from each sheet of a spreadsheet file. These sheets are selected based on the contents of a sheet 'Datasets', with a column called 'Dataset Number', containing numbers identifying the dataset sheets to be read in. In the second column there must be a grouping variable, which will often be named 'Soil'. Optionally, time normalization factors can be given in columns named 'Temperature' and 'Moisture'.
read_spreadsheet(
path,
valid_datasets = "all",
parent_only = FALSE,
normalize = TRUE
)Absolute or relative path to the spreadsheet file
Optional numeric index of the valid datasets, default is to use all datasets
Should only the parent data be used?
Should the time scale be normalized using temperature and moisture normalisation factors in the sheet 'Datasets'?
There must be a sheet 'Compounds', with columns 'Name' and 'Acronym'. The first row read after the header read in from this sheet is assumed to contain name and acronym of the parent compound.
The dataset sheets should be named using the dataset numbers read in from the 'Datasets' sheet, i.e. '1', '2', ... . In each dataset sheet, the name of the observed variable (e.g. the acronym of the parent compound or one of its transformation products) should be in the first column, the time values should be in the second colum, and the observed value in the third column.
In case relevant covariate data are available, they should be given in a sheet 'Covariates', containing one line for each value of the grouping variable specified in 'Datasets'. These values should be in the first column and the column must have the same name as the second column in 'Datasets'. Covariates will be read in from columns four and higher. Their names should preferably not contain special characters like spaces, so they can be easily used for specifying covariate models.
A similar data structure is defined as the R6 class mkindsg, but is probably more complicated to use.