Knowledge Base

Resampling

Why Resampling?

Event data and external data are not likely to have the same sample rate. Also, for the event log, it is unlikely that the timestamps are equidistant. Therefore, if you export the event log together with external data, exporting them at the sample rate of the external dataset means that the event log must be resampled. If you have different external datasets, these may differ in sample rate. If you export all external datasets linked to an observation and these differ in sample rate, some external datasets will be resampled.

How Point Events Are Resampled

When you export event data and external data and choose to export point events to separate columns, Subject–Behavior group combinations that have the same time-stamp after resampling are each written to their own column. When you export to one column, the same Subject–Behavior group combinations that have the same time-stamp after resampling are exported to the same column.

• If you have mainly defined point events (events without duration) in your coding scheme to score and analyze your data, you should export point events to separate columns. As a result, point events that have the same time-stamp after resampling are written to separate columns.
• Different Subject–Behavior group combinations are always exported to separate columns.

How External Data Are Resampled

When exporting event data and external data at a chosen sample rate, any external data recorded at a different sample rate must be resampled. This is done by linearly interpolating the data at the sample time points for which data is missing. For example, if you have heart rate data sampled at 2 Hz and FaceReader data sampled at 29.981 Hz and you export at 29.981 Hz, the heart rate data is linearly interpolated to fill in the missing sample points.

How Event Data And External Data Are Resampled

When both event data and external data are exported together, all data are resampled to the selected sample rate before export. The column with relative time in the export file contains the time-stamps based on the chosen sample rate. Point events without duration are resampled at the first sample following the original time the event occurred. If the One column option was selected, events that occur at the same time are exported to the same column, separated by a comma.

How Missing Values Are Resampled

If your dataset contains missing samples, the missing sample is converted to the symbol NaN. When such external data are exported, the resampled samples will get a NaN value in the export file when at least one of the samples used for interpolation has a NaN value.

Export Data To Text Files With The Find Function

Aim

To quickly export your data by copying and pasting them into another program, or to export a specific selection of your events. Use this method for a quick view of your events or a selection of events.

Procedure

1. Open an observation.
2. Do one of the following:
   • Click the Find button on the toolbar.
   • Choose Edit > Find.
   • Press Ctrl+F.
3. Specify the selection in the upper part of the screen.
4. Click Find.
5. Optionally, hide columns that are not of interest. To select which columns to show or hide, right-click a header and specify your selection.
6. If you scored start-stop events and do not want to export the stop events, select Hide Stop events.
7. To export all events, click the Export all events button. Otherwise, select the events you want to export and click the Export selected events button. You can export the events as a text file or to Excel.

Notes

• The Find function does not create a subset of data ready to be analyzed (for example, to calculate statistics). For this purpose, select data in the Data Profile.
• You can also copy the events and paste them into another program. To do so, right-click one of the cells and select Copy all.
• To use the exported data in a statistical package you may need the event start time together with the event durations. Make sure the Duration column is visible. If it is hidden, right-click a column header, select Show column, and select Duration.

Import Observational Data

Aim

To import observational data into an existing project. Observational data can be:

• Exported The Observer XT data files (*.odx)
• Observer log files from FaceReader (*.odx)
• Exported Viso sessions (*.odx) — see Import ODX Files
• Other data in text format containing a header, a column with time stamps, and one or more columns with codes — see Import Other Observational Data