I'm trying to calculate DSI scores between male-females dyads using the socialindices by Christof Neumann package. I have three dataframes: one containing the behavioural observations, one the observation time and another the presence file. Here are their structures:
> str(observations)
tibble [476 × 6] (S3: tbl_df/tbl/data.frame)
$ date : Date[1:476], format: "2022-10-20" "2022-10-14" "2022-10-14" "2022-09-20" ...
$ focal : chr [1:476] "Sho" "Sho" "Sho" "Buk" ...
$ actor : chr [1:476] "Ginq" "Sho" "Ndaw" "Ginq" ...
$ receiver: chr [1:476] "Sho" "Ndaw" "Sho" "Buk" ...
$ beh : chr [1:476] "Approach" "Approach" "Groom.focal" "Contact.focal" ...
$ dur : int [1:476] NA NA 128 661 223 899 434 NA 19 NA ...
> str(ot)
'data.frame': 209 obs. of 3 variables:
$ focal: chr "Buk" "Buk" "Buk" "Buk" ...
$ date : chr "2022-06-07" "2022-06-11" "2022-06-15" "2022-06-17" ...
$ OT : int 1204 1212 1212 1200 1200 1202 1205 635 1204 963 ...
> str(pres)
'data.frame': 374 obs. of 5 variables:
$ date: chr "2022-01-03" "2022-01-04" "2022-01-05" "2022-01-06" ...
$ Buk : int 1 1 1 0 1 1 1 1 1 1 ...
$ Ginq: int 1 1 1 1 1 1 1 1 1 1 ...
$ Ndaw: int 1 1 1 1 1 1 1 1 1 1 ...
$ Sho : int 1 1 1 1 1 1 1 1 1 1 ...
Since I am interested in seeing how male-female relationships evolve over time, I have chosen to have my males as my focal individuals. That means that in this particular group, I have two focal individuals (Buk and Sho), and three potential partners for each.
This is the code that I'd like to run:
scores <- DSI(observations, ot.source = ot, presence = pres, onlyfocaldyads = F, limit2focalnonfocal = T, duration.NA.treatm = "count")
But I keep getting this error:
Error in strptime(xx, ff, tz = "GMT") : input string is too long
All my date columns are dates and their range is the same. The same calculations worked before when I used all observations, but that doesn't really serve my purpose. If anyone has any idea why this is not working, I'd be very grateful!
DSI() only accepts data.frame() objects, and your observations object is a tibble(). Below is a reprex that generates your error, followed by a working solution:
# install.packages("remotes")
remotes::install_github("gobbios/socialindices")
library(socialindices2)
# Load sample data
data(dataset3)
# Create representative dataframes
observations <- tibble(dataset3[[1]])
ot <- data.frame(dataset3[[2]])
pres <- data.frame(dataset3[[3]])
# Caculate dyadic CSI
scores <- DSI(observations,
ot.source = ot,
presence = pres,
onlyfocaldyads = F,
limit2focalnonfocal = T,
duration.NA.treatm = "count")
# Error in strptime(xx, ff, tz = "GMT") : input string is too long
# Convert observations to df
observations <- data.frame(observations)
# Caculate dyadic CSI again
scores <- DSI(observations,
ot.source = ot,
presence = pres,
onlyfocaldyads = F,
limit2focalnonfocal = T,
duration.NA.treatm = "count")
scores
# i1 i2 type dyad dot cores appr gro prox supp appr.rt gro.rt prox.rt supp.rt DSI zDSI
# 1 a d FNF a_@_d 3877 119 1 2190 1805 1 0.6675996 5.2571514 4.3288838 0.7880743 2.76043 1.17610
# 2 a n FNF a_@_n 3208 98 0 325 192 0 0.0000000 0.9428686 0.5564952 0.0000000 0.37484 -0.56118
# 3 a v FNF a_@_v 3877 119 0 117 67 2 0.0000000 0.2808615 0.1606843 1.5761486 0.50442 -0.38591
# 4 d f FNF d_@_f 6658 200 2 87 195 3 0.7774958 0.1216123 0.2723236 1.3767035 0.63703 -0.24598
# 5 f n FNF f_@_n 6026 179 2 580 421 2 0.8590387 0.8957793 0.6496022 1.0140604 0.85462 -0.11512
# 6 f v FNF f_@_v 6658 200 7 342 569 0 2.7212355 0.4780623 0.7946264 0.0000000 0.99848 0.07002
# 7 d j FNF d_@_j 4162 119 0 92 459 1 0.0000000 0.2057254 1.0254279 0.7341096 0.49132 -0.41839
# 8 j n FNF j_@_n 3481 98 2 96 345 4 1.4870920 0.2566666 0.9215303 3.5109037 1.54405 0.55217
# 9 j v FNF j_@_v 4162 119 4 251 130 0 2.4875384 0.5612726 0.2904262 0.0000000 0.83481 -0.07171