Different ways to summarize phases

[Sarah Martini]

Hello all!

I am trying to understand the difference between several methods people apply to summarize phase beginning/end dates and lengths.

With phase length – I was wondering whether it is more appropriate to apply the Span() or Interval() function in Oxcal (within a phase model with boundaries). From the manual, it seems like Interval() includes potential not directly dated events within the phase, whereas Span() draws just from dated events. I mainly see people use Span() in published articles though. Can someone explain the difference?

Regarding summarizing phase boundaries – I recently ran into the Manning and Hart 2019 paper where they use a Date() function to summarize the phase beginning/end. I can’t seem to find this function in the online manual. Instead, I have used and reported Boundary ranges. Does the Date() function still exist and have others used it in this way?

Thanks for the help!

Sarah

--

Sarah Martini (she/her)

[Erik Marsh]

Hi Sarah – 

1. Check out Andrew's answer in this post. In my experience, people tend to use them interchangeably, even though they are slightly different. In many cases, Interval is perhaps the more conservative choice, which is reflected in its larger error ranges.

2. Date() is an incorrect way to summarize a phase. As Christopher puts it here (see also the rest of the thread), Date() "is certainly no guide to the length of the phase". Instead, it's an "estimate for the date of an undated [event] within that phase." I usually use Date() when I have other priors, not just a phase. For example, if I have a depth sequence, I place Date() at a specific depth and get an estimate for an undated sample at that depth.

3. As for phase beginning and end, best practice is what you're doing, referring to starting and ending boundaries. Others use the queries First and Last (the first and last events in a phase – see the manual), which are often similar to, but not the same, as phase boundaries.

The best way to summarize a phase, as far as I can tell, is in Bronk Ramsey (2017:Fig 3) – on the same line, the two boundaries (green and red), the KDE in the middle, and the medians below. You can visualize this with View > Plot Stacks.

Hope this helps, Erik

[Sarah Martini]

Hi Erik!

Thanks for the clarifications - I had missed those previous posts!

I definitely feel more confident in the modeling choices that I was making now. And good to know that there is no set answer for the Span() vs. Interval () debate - I had wondered that for years!

Sarah

[Christopher Ramsey]

Span and interval are often similar:

Span is the span of the events actually given in the model
Interval is the interval between the start and end of the phase and so includes events within the phase not actually included in the model

The interval is always longer than the span but the differences become small if there are many events included within the model.

Best wishes

Christopher

> --
> You received this message because you are subscribed to the Google Groups "OxCal" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to oxcal+un...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/oxcal/5d96f90d-b7f9-4abc-a1d6-9a9487236c1dn%40googlegroups.com.