Hi list members,
I'm doing some analysis about differences in behaviours between rural and
urban birds and, after reading and searching in different sources, I have a lot
of doubts about how I'm performing them. I would greatly appreciate any
feedback from you. Here are my questions, models and results:
Background
I performed some behavioural test on individuals belonging to the same
territory (breeding birds), some of them located in rural areas and others in
urban ones. I have
7 variables describing different behaviours in 178 breeding birds, most of them
sharing territories as they are mates. Some of these variables should be
considered as censored data.
My main questions are: 1) which is the relationship between these behaviours
and 2) whether urban birds differ in these behaviours (means) and/or in the
strength of their relationships compare with rural ones.
######################################################################
MODELS AND RESULTS
QUESTION 1) which is the relationship between the behaviours measured:
prior = list(R = list(V = diag(7), nu = 8), G = list(G1 = list(V = diag(7), nu
= 8)))
m1 <- MCMCglmm(fixed = cbind (var1, var2min, var2max, var3min, var3max,
var4min, var4max, var5, var6min, var6max, var7) ~ trait - 1, random = ~
us(trait):nest, rcov = ~ us(trait):units, prior = prior, family
=c("gaussian", "cengaussian", "cengaussian",
"cengaussian", "poisson", "cengaussian",
"poisson"), nitt = 60000, burnin = 1000, thin = 25, data = datos)
I obtained the correlation between behaviours with the general formula:
model$VCV[,"var1:var2.nest"]/sqrt(model$VCV[,"var1:var1.nest"]*model$VCV[,"var2:var2.nest"]))
so that:
cor(var2:var1): 0.379; 95%CI = 0.376 - 0.383
cor(var3:var1): 0.246; 95%CI = 0.242 - 0.249
cor(var4:var1): 0.150; 95%CI = 0.146 - 0.155
cor(var5:var1): -0.022; 95%CI = -0.027 - -0.017
cor(var6:var1): 0.171; 95%CI = 0.167 - 0.176
cor(var7:var1): 0.001; 95%CI = -0.004 - 0.006
cor(var3:var2): 0.364; 95%CI = 0.360 - 0.369
cor(var4:var2): 0.121; 95%CI = 0.115 - 0.127
cor(var5:var2): -0.037; 95%CI = -0.044 - -0.030
cor(var6:var2): 0.209; 95%CI = 0.203 - 0.215
cor(var7:var2): -0.031; 95%CI = -0.038 - -0.024
cor(var4:var3): 0.062; 95%CI = 0.056 - 0.068
cor(var5:var3): 0.037; 95%CI = 0.030 - 0.045
cor(var6:var3): 0.210; 95%CI = 0.204 - 0.216
cor(var7:var3): 0.028; 95%CI = 0.021 - 0.035
cor(var5:var4): -0.436; 95%CI = -0.442 - -0.429
cor(var6:var4): 0.133; 95%CI = 0.126 - 0.140
cor(var7:var4): -0.120; 95%CI = -0.128 - -0.111
cor(var6:var5): -0.160; 95%CI = -0.169 - -0.151
cor(var7:var5): 0.346; 95%CI = 0.336 - 0.357
cor(var7:var6): -0.371; 95%CI = -0.379 - -0.364
So my first response would be that yes, all my behavioural measures are
correlated (with different strength or sign). Just to be sure: even if birds are
nested within territories (nest), these correlations are at the individual level
(within the individual), and nest is a random term because we replicate
individuals within the same territory, but nothing about resemblance between
mates. OK?
######################################################################
QUESTION 2) urban birds differ in mean or the strength of the relationship
between these behaviours compare with rural ones. I include in models the term
"habitat" which is a factor with 2 levels (urban or rural).
Here I have some doubts, as I'm not sure how to do the model:
m2a <- MCMCglmm(fixed = cbind(var1, var2min, var2max, var3min, var3max,
var4min, var4max, var5, var6min, var6max, var7) ~ trait - 1 + habitat, random =
~ us(trait):nest, rcov = ~ us(trait):units, prior = prior,family =
c("gaussian", "cengaussian", "cengaussian",
"cengaussian", "poisson", "cengaussian",
"poisson"), nitt = 60000, burnin = 1000, thin = 25, data = datos)
m2b <- MCMCglmm(fixed = cbind(var1, var2min, var2max, var3min, var3max,
var4min, var4max, var5, var6min, var6max, var7) ~ trait - 1 + trait:habitat,
random = ~ us(trait):nest, rcov = ~ us(trait):units, prior = prior,family =
c("gaussian", "cengaussian", "cengaussian",
"cengaussian", "poisson", "cengaussian",
"poisson"), nitt = 60000, burnin = 1000, thin = 25, data = datos)
m2a, and m2b are different models, but I'm not sure which is their meanings:
after reading, what I understood is that m2a test the hypothesis that the
relationship
between variables changes but in the same way between habitats, while in m2b the
idea is that habitat type affect the relationship between variables differently.
DIC(m2a): 1514.612
DIC(m2b): 1517.572
m2a is the best model, but m2b is close (∆DIC= 2.96), so should I conclude that
the relationship between variables is similar in both habitat types?
Then, I don't know how to obtain the correlations between the different
behaviours using this model (m2b). I find a recomendation in the Rlist,
something like this:
m2c <- MCMCglmm(fixed = cbind(var1, var2min, var2max, var3min, var3max,
var4min, var4max, var5, var6min, var6max, var7) ~ trait - 1 +
trait:habitat,random = ~ us(trait:at.level(habitat, 1)):nest +
us(trait:at.level(habitat, 2)):nest, rcov = ~ us(trait):units, prior = list(R =
list(V = diag(7), nu = 8), G = list(G1 = list(V = diag(7), nu = 8), G2 = list(V
= diag(7), nu = 8))), family = c("gaussian", "cengaussian",
"cengaussian", "cengaussian", "poisson",
"cengaussian", "poisson"), nitt = 60000, burnin = 1000, thin
= 25, data = datos)
summary(m2c)
Iterations = 1001:59976
Thinning interval = 25
Sample size = 2360
DIC(m2c): 1566.367
G-structure: ~us(trait:at.level(habitat, 1)):nest
post.mean l-95% CI u-95% CI eff.samp
var1:at.level(habitat, 1):var1:at.level(habitat, 1).nest 0.136304
0.09799 0.17641 2360.00
var2:at.level(habitat, 1):var1:at.level(habitat, 1).nest 0.015494
-0.03910 0.06556 1774.69
var3:at.level(habitat, 1):var1:at.level(habitat, 1).nest 0.002703
-0.06701 0.07044 1596.13
var4:at.level(habitat, 1):var1:at.level(habitat, 1).nest 0.024925
-0.05294 0.11054 1532.04
var5:at.level(habitat, 1):var1:at.level(habitat, 1).nest -0.031660
-0.19513 0.12782 1915.77
var6:at.level(habitat, 1):var1:at.level(habitat, 1).nest 0.008893
-0.09836 0.11516 1174.13
var7:at.level(habitat, 1):var1:at.level(habitat, 1).nest -0.020421
-0.18110 0.14512 1425.96
var1:at.level(habitat, 1):var2:at.level(habitat, 1).nest 0.015494
-0.03910 0.06556 1774.69
var2:at.level(habitat, 1):var2:at.level(habitat, 1).nest 0.410275 0.26237
0.59275 934.78
var3:at.level(habitat, 1):var2:at.level(habitat, 1).nest 0.057976 -0.12494
0.23465 330.66
var4:at.level(habitat, 1):var2:at.level(habitat, 1).nest -0.003364 -0.22167
0.19782 772.17
var5:at.level(habitat, 1):var2:at.level(habitat, 1).nest -0.056700 -0.45976
0.39309 612.34
var6:at.level(habitat, 1):var2:at.level(habitat, 1).nest -0.021934 -0.30452
0.25433 371.81
var7:at.level(habitat, 1):var2:at.level(habitat, 1).nest -0.017623 -0.49615
0.39792 548.59
var1:at.level(habitat, 1):var3:at.level(habitat, 1).nest 0.002703
-0.06701 0.07044 1596.13
var2:at.level(habitat, 1):var3:at.level(habitat, 1).nest 0.057976 -0.12494
0.23465 330.66
var3:at.level(habitat, 1):var3:at.level(habitat, 1).nest 0.642849 0.34019
0.98322 305.33
var4:at.level(habitat, 1):var3:at.level(habitat, 1).nest -0.054583 -0.33766
0.21244 623.97
var5:at.level(habitat, 1):var3:at.level(habitat, 1).nest 0.108017 -0.45137
0.68393 513.78
var6:at.level(habitat, 1):var3:at.level(habitat, 1).nest -0.005969 -0.40265
0.35762 368.33
var7:at.level(habitat, 1):var3:at.level(habitat, 1).nest 0.047296 -0.54854
0.64387 425.52
var1:at.level(habitat, 1):var4:at.level(habitat, 1).nest 0.024925
-0.05294 0.11054 1532.04
var2:at.level(habitat, 1):var4:at.level(habitat, 1).nest -0.003364 -0.22167
0.19782 772.17
var3:at.level(habitat, 1):var4:at.level(habitat, 1).nest -0.054583 -0.33766
0.21244 623.97
var4:at.level(habitat, 1):var4:at.level(habitat, 1).nest 0.838939 0.33055
1.51471 356.01
var5:at.level(habitat, 1):var4:at.level(habitat, 1).nest -0.993110 -2.24249
-0.01864 263.03
var6:at.level(habitat, 1):var4:at.level(habitat, 1).nest 0.336994 -0.23357
1.00589 181.84
var7:at.level(habitat, 1):var4:at.level(habitat, 1).nest -0.735804 -2.00597
0.11626 165.63
var1:at.level(habitat, 1):var5:at.level(habitat, 1).nest -0.031660
-0.19513 0.12782 1915.77
var2:at.level(habitat, 1):var5:at.level(habitat, 1).nest -0.056700 -0.45976
0.39309 612.34
var3:at.level(habitat, 1):var5:at.level(habitat, 1).nest 0.108017 -0.45137
0.68393 513.78
var4:at.level(habitat, 1):var5:at.level(habitat, 1).nest -0.993110 -2.24249
-0.01864 263.03
var5:at.level(habitat, 1):var5:at.level(habitat, 1).nest 2.948040 0.50599
6.20484 213.50
var6:at.level(habitat, 1):var5:at.level(habitat, 1).nest -0.878025 -2.40912
0.36680 157.95
var7:at.level(habitat, 1):var5:at.level(habitat, 1).nest 1.928700 -0.45261
4.73825 105.74
var1:at.level(habitat, 1):var6:at.level(habitat, 1).nest 0.008893
-0.09836 0.11516 1174.13
var2:at.level(habitat, 1):var6:at.level(habitat, 1).nest -0.021934 -0.30452
0.25433 371.81
var3:at.level(habitat, 1):var6:at.level(habitat, 1).nest -0.005969 -0.40265
0.35762 368.33
var4:at.level(habitat, 1):var6:at.level(habitat, 1).nest 0.336994 -0.23357
1.00589 181.84
var5:at.level(habitat, 1):var6:at.level(habitat, 1).nest -0.878025 -2.40912
0.36680 157.95
var6:at.level(habitat, 1):var6:at.level(habitat, 1).nest 1.283158 0.40113
2.59803 124.27
var7:at.level(habitat, 1):var6:at.level(habitat, 1).nest -1.125402 -2.76629
0.06360 155.35
var1:at.level(habitat, 1):var7:at.level(habitat, 1).nest -0.020421
-0.18110 0.14512 1425.96
var2:at.level(habitat, 1):var7:at.level(habitat, 1).nest -0.017623 -0.49615
0.39792 548.59
var3:at.level(habitat, 1):var7:at.level(habitat, 1).nest 0.047296 -0.54854
0.64387 425.52
var4:at.level(habitat, 1):var7:at.level(habitat, 1).nest -0.735804 -2.00597
0.11626 165.63
var5:at.level(habitat, 1):var7:at.level(habitat, 1).nest 1.928700 -0.45261
4.73825 105.74
var6:at.level(habitat, 1):var7:at.level(habitat, 1).nest -1.125402 -2.76629
0.06360 155.35
var7:at.level(habitat, 1):var7:at.level(habitat, 1).nest 2.999515 0.57516
6.63337 80.53
~us(trait:at.level(habitat, 2)):nest
post.mean l-95% CI u-95% CI eff.samp
var1:at.level(habitat, 2):var1:at.level(habitat, 2).nest 0.1656123
0.11786 0.22106 2360.0
var2:at.level(habitat, 2):var1:at.level(habitat, 2).nest 0.0334815
-0.02249 0.09568 1884.9
var3:at.level(habitat, 2):var1:at.level(habitat, 2).nest 0.0276997
-0.03648 0.10608 1703.1
var4:at.level(habitat, 2):var1:at.level(habitat, 2).nest 0.0094879
-0.06105 0.08723 1740.3
var5:at.level(habitat, 2):var1:at.level(habitat, 2).nest 0.0073708
-0.10499 0.12369 2054.9
var6:at.level(habitat, 2):var1:at.level(habitat, 2).nest 0.0210704
-0.06467 0.10751 1239.8
var7:at.level(habitat, 2):var1:at.level(habitat, 2).nest -0.0225721
-0.17452 0.12613 1311.9
var1:at.level(habitat, 2):var2:at.level(habitat, 2).nest 0.0334815
-0.02249 0.09568 1884.9
var2:at.level(habitat, 2):var2:at.level(habitat, 2).nest 0.3486283
0.22524 0.50847 1254.3
var3:at.level(habitat, 2):var2:at.level(habitat, 2).nest 0.0893576
-0.02938 0.22749 1103.3
var4:at.level(habitat, 2):var2:at.level(habitat, 2).nest 0.0170261 -0.12427
0.15590 1276.6
var5:at.level(habitat, 2):var2:at.level(habitat, 2).nest 0.0204992
-0.19159 0.26654 1316.1
var6:at.level(habitat, 2):var2:at.level(habitat, 2).nest 0.0539765 -0.09843
0.21485 1039.3
var7:at.level(habitat, 2):var2:at.level(habitat, 2).nest -0.0663682 -0.33668
0.22229 842.1
var1:at.level(habitat, 2):var3:at.level(habitat, 2).nest 0.0276997
-0.03648 0.10608 1703.1
var2:at.level(habitat, 2):var3:at.level(habitat, 2).nest 0.0893576
-0.02938 0.22749 1103.3
var3:at.level(habitat, 2):var3:at.level(habitat, 2).nest 0.4424986
0.24772 0.65768 958.6
var4:at.level(habitat, 2):var3:at.level(habitat, 2).nest 0.0209836 -0.15302
0.18846 1117.1
var5:at.level(habitat, 2):var3:at.level(habitat, 2).nest 0.0226774
-0.26750 0.30541 1367.4
var6:at.level(habitat, 2):var3:at.level(habitat, 2).nest 0.0731457 -0.12227
0.28489 883.4
var7:at.level(habitat, 2):var3:at.level(habitat, 2).nest -0.0251657 -0.36957
0.31977 755.0
var1:at.level(habitat, 2):var4:at.level(habitat, 2).nest 0.0094879
-0.06105 0.08723 1740.3
var2:at.level(habitat, 2):var4:at.level(habitat, 2).nest 0.0170261 -0.12427
0.15590 1276.6
var3:at.level(habitat, 2):var4:at.level(habitat, 2).nest 0.0209836 -0.15302
0.18846 1117.1
var4:at.level(habitat, 2):var4:at.level(habitat, 2).nest 0.4854582 0.23937
0.76309 1100.7
var5:at.level(habitat, 2):var4:at.level(habitat, 2).nest -0.1886537 -0.56073
0.10975 930.3
var6:at.level(habitat, 2):var4:at.level(habitat, 2).nest -0.0009642 -0.22502
0.21617 968.9
var7:at.level(habitat, 2):var4:at.level(habitat, 2).nest 0.0949310 -0.36272
0.50778 760.9
var1:at.level(habitat, 2):var5:at.level(habitat, 2).nest 0.0073708
-0.10499 0.12369 2054.9
var2:at.level(habitat, 2):var5:at.level(habitat, 2).nest 0.0204992
-0.19159 0.26654 1316.1
var3:at.level(habitat, 2):var5:at.level(habitat, 2).nest 0.0226774
-0.26750 0.30541 1367.4
var4:at.level(habitat, 2):var5:at.level(habitat, 2).nest -0.1886537 -0.56073
0.10975 930.3
var5:at.level(habitat, 2):var5:at.level(habitat, 2).nest 1.0214578
0.38161 1.84704 756.4
var6:at.level(habitat, 2):var5:at.level(habitat, 2).nest 0.0261338 -0.36384
0.41970 1040.3
var7:at.level(habitat, 2):var5:at.level(habitat, 2).nest 0.0375229 -0.70157
0.93888 704.7
var1:at.level(habitat, 2):var6:at.level(habitat, 2).nest 0.0210704
-0.06467 0.10751 1239.8
var2:at.level(habitat, 2):var6:at.level(habitat, 2).nest 0.0539765 -0.09843
0.21485 1039.3
var3:at.level(habitat, 2):var6:at.level(habitat, 2).nest 0.0731457 -0.12227
0.28489 883.4
var4:at.level(habitat, 2):var6:at.level(habitat, 2).nest -0.0009642 -0.22502
0.21617 968.9
var5:at.level(habitat, 2):var6:at.level(habitat, 2).nest 0.0261338 -0.36384
0.41970 1040.3
var6:at.level(habitat, 2):var6:at.level(habitat, 2).nest 0.6139191 0.28345
1.00710 784.2
var7:at.level(habitat, 2):var6:at.level(habitat, 2).nest -0.2443531 -0.82849
0.18576 816.9
var1:at.level(habitat, 2):var7:at.level(habitat, 2).nest -0.0225721
-0.17452 0.12613 1311.9
var2:at.level(habitat, 2):var7:at.level(habitat, 2).nest -0.0663682 -0.33668
0.22229 842.1
var3:at.level(habitat, 2):var7:at.level(habitat, 2).nest -0.0251657 -0.36957
0.31977 755.0
var4:at.level(habitat, 2):var7:at.level(habitat, 2).nest 0.0949310 -0.36272
0.50778 760.9
var5:at.level(habitat, 2):var7:at.level(habitat, 2).nest 0.0375229 -0.70157
0.93888 704.7
var6:at.level(habitat, 2):var7:at.level(habitat, 2).nest -0.2443531 -0.82849
0.18576 816.9
var7:at.level(habitat, 2):var7:at.level(habitat, 2).nest 1.6442687 0.44836
3.38122 288.9
R-structure: ~us(trait):units
post.mean l-95% CI u-95% CI eff.samp
var1:var1.units 0.062164 0.051459 0.07416 2360.0
var2:var1.units 0.006561 -0.008157 0.02402 1853.0
var3:var1.units 0.004655 -0.015026 0.02618 1677.4
var4:var1.units 0.009616 -0.024670 0.04814 1686.9
var5:var1.units -0.020257 -0.099360 0.06894 1409.0
var6:var1.units 0.009687 -0.024687 0.04640 2162.1
var7:var1.units -0.018389 -0.088376 0.04871 1997.5
var1:var2.units 0.006561 -0.008157 0.02402 1853.0
var2:var2.units 0.171884 0.133601 0.22208 1575.2
var3:var2.units 0.011132 -0.041338 0.06197 629.9
var4:var2.units -0.032961 -0.106789 0.03922 1463.1
var5:var2.units 0.029247 -0.141493 0.20121 1366.6
var6:var2.units -0.019018 -0.102393 0.06462 1130.2
var7:var2.units 0.025949 -0.131932 0.19386 1098.0
var1:var3.units 0.004655 -0.015026 0.02618 1677.4
var2:var3.units 0.011132 -0.041338 0.06197 629.9
var3:var3.units 0.297220 0.194598 0.41320 656.3
var4:var3.units -0.003164 -0.171143 0.15681 384.9
var5:var3.units 0.024245 -0.378239 0.40849 361.4
var6:var3.units 0.011216 -0.140303 0.19338 376.0
var7:var3.units 0.002151 -0.367842 0.33249 362.4
var1:var4.units 0.009616 -0.024670 0.04814 1686.9
var2:var4.units -0.032961 -0.106789 0.03922 1463.1
var3:var4.units -0.003164 -0.171143 0.15681 384.9
var4:var4.units 0.865977 0.555265 1.23769 691.4
var5:var4.units -1.616190 -2.369540 -0.97822 540.4
var6:var4.units 0.275559 0.043621 0.52288 780.3
var7:var4.units -0.620830 -1.223925 -0.11597 420.1
var1:var5.units -0.020257 -0.099360 0.06894 1409.0
var2:var5.units 0.029247 -0.141493 0.20121 1366.6
var3:var5.units 0.024245 -0.378239 0.40849 361.4
var4:var5.units -1.616190 -2.369540 -0.97822 540.4
var5:var5.units 4.603627 2.611698 6.65959 245.5
var6:var5.units -0.795076 -1.442050 -0.24446 545.7
var7:var5.units 1.954516 0.586883 3.49957 301.4
var1:var6.units 0.009687 -0.024687 0.04640 2162.1
var2:var6.units -0.019018 -0.102393 0.06462 1130.2
var3:var6.units 0.011216 -0.140303 0.19338 376.0
var4:var6.units 0.275559 0.043621 0.52288 780.3
var5:var6.units -0.795076 -1.442050 -0.24446 545.7
var6:var6.units 0.857967 0.522307 1.27700 458.0
var7:var6.units -1.067046 -1.813576 -0.52592 356.9
var1:var7.units -0.018389 -0.088376 0.04871 1997.5
var2:var7.units 0.025949 -0.131932 0.19386 1098.0
var3:var7.units 0.002151 -0.367842 0.33249 362.4
var4:var7.units -0.620830 -1.223925 -0.11597 420.1
var5:var7.units 1.954516 0.586883 3.49957 301.4
var6:var7.units -1.067046 -1.813576 -0.52592 356.9
var7:var7.units 3.018702 1.379693 5.09078 205.3
Location effects: cbind(var1, var2, logcaprmax, var3, logtcaemax, var4,
logzorromax, var5, var6, loghalconmax, var7) ~ trait - 1 + trait:habitat
post.mean l-95% CI u-95% CI eff.samp pMCMC
traitvar1 1.68859 1.59946 1.76748 2360.0 < 4e-04 ***
traitvar2 1.11222 0.94128 1.27189 1142.8 < 4e-04 ***
traitvar3 1.30170 1.06826 1.59399 290.6 < 4e-04 ***
traitvar4 1.26373 0.87937 1.62822 294.9 < 4e-04 ***
traitvar5 -0.18299 -0.97919 0.56381 353.9 0.66441
traitvar6 1.48153 0.89085 2.01253 120.3 < 4e-04 ***
traitvar7 -1.08708 -2.03436 -0.19270 127.1 0.01949 *
traitvar1:habitat1 -0.48558 -0.60980 -0.36445 1922.9 < 4e-04 ***
traitvar2:habitat1 -0.66787 -0.89365 -0.43531 1141.6 < 4e-04 ***
traitvar3:habitat1 -0.48361 -0.82642 -0.15277 437.0 0.00847 **
traitvar4:habitat1 -0.29592 -0.79364 0.16455 401.0 0.22797
traitvar5:habitat1 0.06428 -0.91838 1.02105 500.2 0.90254
traitvar6:habitat1 -0.55078 -1.21994 0.10046 175.9 0.10000 .
traitvar7:habitat1 -0.44890 -1.67698 0.75958 189.2 0.45932
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
and correlations for the variables for the two levels of habitat are:
habitat = RURAL
var2:var1: 0.06581789; 95% CI = 0.06133871 - 0.07029706
var3:var1: 0.008656465; 95% CI = 0.00392139 - 0.01339154
var4:var1: 0.07288664; 95% CI = 0.06811906 - 0.07765422
var5:var1: -0.04830374; 95% CI = -0.05330634 - -0.04330114
var6:var1: 0.02026321; 95% CI = 0.01521869 - 0.02530773
var7:var1: -0.03021775; 95% CI = -0.03528822 - -0.02514727
var3:var2: 0.1111226; 95% CI = 0.1044198 - 0.1178254
var4:var2: -0.004587257; 95% CI = -0.01146466 - 0.002290146
var5:var2: -0.05334263; 95% CI = -0.06069082 - -0.04599444
var6:var2: -0.02931499; 95% CI = -0.03688508 - -0.0217449
var7:var2: -0.01759684; 95% CI = -0.02531301 - -0.00988067
var4:var3: -0.07251074; 95% CI = -0.07990936 - -0.06511211
var5:var3: 0.07728087; 95% CI = 0.06903799 - 0.08552375
var6:var3: -0.006868517; 95% CI = -0.01497649 - 0.001239459
var7:var3: 0.03703707; 95% CI = 0.02839657 - 0.04567758
var5:var4: -0.5945241; 95% CI = -0.601466 - -0.5875823
var6:var4: 0.2954133; 95% CI = 0.2859855 - 0.3048411
var7:var4: -0.43075; 95% CI = -0.4396149 - -0.421885
var4:var5: -0.5945241; 95% CI = -0.601466 - -0.5875823
var6:var5: -0.4162762; 95% CI = -0.4261684-0.406384
var7:var5: 0.6000609; 95% CI = 0.590975 - 0.6091469
var7:var6: -0.5416963; 95% CI = -0.5498123 - -0.5335803
habitat = URBAN
var2:var1: 0.1381003; 95% CI = 0.1332987 - 0.1429019
var3:var1: 0.1009115; 95% CI = 0.09578722 - 0.1060357
var4:var1: 0.03348612; 95% CI = 0.02817647 - 0.03879578
var5:var1: 0.0186588; 95% CI = 0.01314655 - 0.02417105
var6:var1: 0.06562708; 95% CI = 0.06023977 - 0.07101439
var7:var1: -0.04184128; 95% CI = -0.04743616 - -0.0362464
var3:var2: 0.2239385; 95% CI = 0.2178998 - 0.2299772
var4:var2: 0.04107495; 95% CI = 0.03427635 - 0.04787355
var5:var2: 0.03300118; 95% CI = 0.02560607 - 0.04039628
var6:var2: 0.1144631; 95% CI = 0.1079228 - 0.1210033
var7:var2: -0.08634689; 95% CI = -0.09352826 - -0.07916552
var4:var3: 0.04332442; 95% CI = 0.03626806 - 0.05038079
var5:var3: 0.03267068; 95% CI = 0.02472427 - 0.04061708
var6:var3: 0.1357245; 95% CI = 0.1285366 - 0.1429123
var7:var3: -0.03101644; 95% CI = -0.03889608 - -0.0231368
var5:var4: -0.2505957; 95% CI = -0.2585397 - -0.2426517
var6:var4: -0.002795438; 95% CI = -0.01065961 - 0.005068738
var7:var4: 0.1007233; 95% CI = 0.09204078 - 0.1094059
var4:var5: -0.2505957; 95% CI = -0.2585397 - -0.2426517
var6:var5: 0.03341757; 95% CI = 0.02415877 - 0.04267638
var7:var5: 0.02453784; 95% CI = 0.01333783 - 0.03573786
var7:var6: -0.2279325; 95% CI = -0.2364717 - -0.2193934
However, the DIC of this model is the worst!
My main question regarding this part of the analysis is how can I know that
habitat significantly affect the correlation between behaviours and which are
the mean values observed for each behaviour in each habitat type. For the first
part of the question, DIC of the m2c model (1566.367) is higher than the DIC of
m1 model (without habitat effect) and m2 (with "trait -1 + habitat" as
fixed effect), but I know that DIC is not equivalent to AIC, so I'm not sure
how much I should trust on it...
I know it's too much, but I would greatly appreciate some feedback from you.
As you can see, my statistical baggage is low and I'm trying to fix that
with manuals and
forum posts which sometimes can be very confusing.
Cheers,
Martina
Dra. Martina Carrete
Dpt Physical, Chemical and Natural Systems
Universidad Pablo de Olavide,
Ctra. Utrera km 1
41013, Sevilla
[[alternative HTML version deleted]]