Process Replicates

The Process Replicates option in the Data menu is the preferred procedure for managing replicates in TableCurve 2D. This option automatically searches a composite data table for replicate X values, and replaces the various instances with a single table entry where the Y value consists of the mean of the individual Y values, and the weight value consists of 1/variance (1/SD²). This option also stores the n for each replicate entry, enabling Error Bars to be expressed as Student-t distribution based confidence or prediction intervals. In TableCurve 2D, replicates are based on the X-values coinciding to within 1E-8 fractional precision.

Saving

If you choose to Save a processed replicate data table, you should do so as an ASCII file in order to preserve the n for each of the entries.

Appending

The recommended approach for replicates in TableCurve 2D is to externally create a composite data set containing the multiple experiments, or to use the Append options within TableCurve 2D to create a single data table containing all pertinent replicates. It is recommended that you save this file as the parent for all subsequent work. This composite table can then be automatically processed for replicates using this option.

If you are appending multiple data sets, you should not invoke this option until you have appended all of the data sets. TableCurve 2D's data table retains only the variance at each X-value, not the number of samples at that X. This means that you cannot meaningfully append a data set to a table that already consists of averaged values since the averaged set and the additional set would be given equal weight in the averaging.

Note that Column-Specified Import does allow the importing of weights as standard deviations, but the n for each entry is not available, and as such, Error Bars associated with replicates are limited to multiples of the standard deviation and to confidence intervals based on a normal distribution. Student-t distribution based confidence and prediction intervals will not be available.

Restoring Original Data

When the replicates are averaged, whether by this option or automatically, the original appended series can be restored with the Reset XY Data option.

Error Bars

Error bars will be available in the Data menu's View Data option if replicates are averaged. If all of the replicates for a given X-value are identical, an SD of 0 is used for the error bar. If only a single value is available, TableCurve 2D uses the conservative approach of setting the SD for that point to the range of the Y-variable.