add suggestion to calculate volume of behaviour

vignettes/chapter12_TimeUseAnalysis.Rmd

@@ -63,6 +63,9 @@ GGIR provides the following metrics over the time windows calculated, i.e., full
 -   Number of bouts: Number of bouts per behavioural class.
 -   Fragmentation: The fragmentation metrics as discussed in the previous chapter. Here no distinction is made between bouted or unbouted behavour. Note that fragmentation classes sometimes group multiple intensity levels, e.g. the fragmentation of physical activity reflects the fragmentation of LIPA and MVPA combined relative to Inactive time.
 
+On a side note - If you multiply Acceleration by Duration for a given class, and by that combine the information from both variables, you would arrive at a volume measure of behaviour. This is similar to the construct of calories over time. I think this would be a much richer way of describing the data as opposed to the conventional approach that only looks at either time spent per behavioural class or average acceleration in an entire day.
+
+
 ### Complementary variables
 
 If your primary interest is on sleep research then we recommend you to work with the GGIR part 4 reports. However, for those who want to look at interactions between behaviour and sleep, GGIR part 5 reports include sleep estimates as used for the part 5 analysis. Note that in part 5 the criteria for sleep estimate inclusion are different than for part 4. In part 5 we are happy with any estimate, even if the accelerometer was not worn during the night.