Specifies the statistic for an `totaldegreeSender` effect in the tie-oriented model or the sender activity rate step of the actor-oriented model.
Usage
totaldegreeSender(scaling = c("none", "prop", "std"), consider_type = TRUE)Arguments
- scaling
the method for scaling the degree statistic. Default is to not scale the statistic (scaling = "none"). Alternatively, scaling of the raw degree counts by two times the number of past events at time t can be requested with 'prop' or standardization of the raw degree counts per time point can be requested with 'std'.
- consider_type
logical, indicates whether to count the degrees separately for each event type (TRUE, default) or sum degrees across different event types (FALSE).
Details
A total degree of the sender effect refers to the tendency for actors to send events if they have send and received more past events. The statistic at timepoint t for dyad (i,j) (tie-oriented model) or sender i (actor-oriented model) is equal to the number of events send and received by actor i before timepoint t. Note that the 'totaldegreeSender' effect is only defined for directed events.
Optionally, a scaling method can be set with scaling. By scaling the
degree count by the total number of past events times two, the statistic
refers to the fraction of past events times two that involved actor i. At
the first time point, when no events did previously occur, it is assumed
that every actor is equally likely to send a message and the statistic is
set equal to 1/n, where n refers to the number of actors.
See also
indegreeSender, indegreeReceiver,
outdegreeSender, outdegreeReceiver, or
totaldegreeReceiver for other types of degree effects.
Examples
effects <- ~ totaldegreeSender()
reh_tie <- remify::remify(history, model = "tie")
remstats(reh = reh_tie, tie_effects = effects)
#> Relational Event Network Statistics
#> > Model: tie-oriented
#> > Computation method: per time point
#> > Dimensions: 115 time points x 90 dyads x 2 statistics
#> > Statistics:
#> >> 1: baseline
#> >> 2: totaldegreeSender
reh_actor <- remify::remify(history, model = "actor")
remstats(reh = reh_actor, sender_effects = effects)
#> Relational Event Network Statistics
#> > Model: actor-oriented
#> > Computation method: per time point
#> > Sender model:
#> >> Dimensions: 115 time points x 10 actors x 2 statistics
#> >> Statistics:
#> >>> 1: baseline
#> >>> 2: totaldegreeSender