Using known fate for nests- formatting data?

[zugunruhe1]

Hello, I am trying to format nest data to use with the known fate model (because I used temperature loggers that allowed me to detect when the nest failed). I’ve read the known fate chapter, but am struggling on how to translate the formatting. For instance, would each interval correspond to a day in the nesting season? If so, that might be problematic because the staggered entry example of data, which applies most closely to my nest data since all my nests were obviously not found on the same day (page 16-13), assumes that the end of the tracking period was the same for all individuals. But I’m thinking the real “end” should not exceed what the fledging date would have been for a particular nest. For instance, if I had a nest that failed in the beginning of the season, I could code it “11” when it failed, but would giving it pairs of 0’s for every day thereafter until the end of the nesting season be telling MARK that it theoretically could have been active over the entire nesting season? I'm particularly concerned about this because I'd like to test if there are any effects of variation in DSR across the nesting season. Is coding failed nests with 0's for the remainder of the nesting season an invalid approach, or am I overthinking this? I searched the MARK forum and found a similar question but the only answer was to use known fate, so I’d really appreciate any specific help on formatting. Thanks so much!

[Jochen]

Hi,
this is exactly what the nest survival model and ist special data format were designed for. No need to struggle with 1 and 0 etc. The book chapter on nest survival gives all necessary information.

[zugunruhe1]

Thanks but my whole problem lies in the fact that the nest survival model assumes that the exact end date is unknown- so I don't know how this solves my problem.The only example the nest survival chapter gives where "date last known alive" and "date last checked" are the same says the interepretation of such a history would be "invalid".

[Rotella]

In the nest-survival model, you can have the last date known alive be the day before the nest failure occurs, in which case you know the date the nest fails, i.e., it fails on the last day of observation. You might try working with a simple input example like the one below with an S(.) model. You'll see that the daily survival rate is 0.90 (nests survived 900 daily trials and there were 100 daily trials that ended in failure). Good luck.

Nest survival group = 1;
1 5 6 1 100;
1 6 6 0 100;

[zugunruhe1]

Thank you very much. I did think of that as the closest approximation to my conditions, but wouldn't that still be inaccurate since in your example MARK would interpret the "real" end date as day 5.5 for the first set of nests? Are you saying the half-day wouldn't make much of a difference, or that if I did it that way for every nest it would be "standardized" across all my nests so it would be an acceptable alternative?

Thank you for the answer and (hopefully) bearing with my follow-up questions, I am not terribly left-brained so I like to make sure I'm interpreting things correctly

[Rotella]

If you want to consider finer details of when the failure occurs within the last day, I think you would need to work with finer intervals of time or move to a time-to-death analysis. There are plenty of good papers on time-to-death approaches. For example, Dennis Heisey, Terry Shaffer, and Gary White wrote a very useful overview in Studies in Avian Biology No. 34:13–33.

[Bryan Hamilton]

How do "known fate" models differ from "time to death" models?