30 
males, and both sexes combined were 51, 52, and 52%, 
respectively; recovery rates of the three groups were 1.6, 
1.6, and 1.4%, respectively. 
Adults and immatures. — Survival and recovery rates for 
adult and immature doves were estimated for 11 of 15 ref- 
erence areas by use of program BROWNIE (Tables A-73 and 
A-74), 
For the entire CMU the unweighted average survival 
rates for adult and immature doves were 52.7 and 43.8 % : 
respectively, Adult doves from New Mexico had the highest 
survival rate (66.4%) and those banded in Arkansas had 
the lowest (35.4%). Immature doves from Nebraska had 
the highest survival rate (67.8%) and those from Missouri 
had the lowest (26.1%). 
The unweighted average recovery rates for adult and 
immature doves banded in the CMU were 1.5 and 2.0%. 
respectively. Adult doves from Missouri had the highest 
recovery rate (2.3%) and those banded in Nebraska and 
lowa had the lowest (0.6%). Immature doves from 
Arkansas had the highest recovery rate (4.5%) and those 
from Nebraska and New Mexico had the lowest (0.8%). 
Sex Related Differences 
Banding and recovery data for adult male and female 
mourning doves from each State were analyzed to test the 
hypothesis that survival and recovery rates of the two 
cohorts were similar. If no differences are detected, then 
data from the two cohorts within each reference area can 
be pooled. 
Differences in survival and recovery rates. — Banding and 
recovery data for adult males and females from each State 
were subjected to Option 1 of program Brownir, This pro- 
gram provides a chi-square contingency test for differences 
in survival and recovery rates between two cohorts, How- 
ever, if differences are detected, it will not distinguish 
whether they are related to survival or recovery rates, or 
both. 
Seven of 15 reference areas had insufficient data for test- 
ing. Bandings and recoveries of adult females were gener- 
ally the weakest. No differences between adult males and 
adult females were detected in the eight reference areas 
tested (Table A-75). By summing individual test results and 
degrees of freedom, a composite test statistic was derived. 
No difference was detected. 
Differences in survival rates. — Differences in survival 
rates between sexes were tested in 5 of 15 reference areas 
(Tables A-70 and A-71) by a 2-tailed Z-test (Brownie et al. 
1978). Data from South Dakota indicated that males sur- 
vived at a lower rate than females (50.0 vs. 59.1%: 
P < 0.05). Data from Missouri indicated males survived 
at a higher rate than females (48.1 vs, 33.2%: P < 0.01). 
No differences were detected in survival rates between 
males and females in Kansas (48.1 vs. 52.8%), Colorado 
(56.9 vs. 57.6%), and North Texas (43.9 vs. 52.4 % ). There 
was no difference in the composite test statistic which con- 
sidered all areas equally. 
Differences in recovery rates. — Differences in recovery 
rates between sexes were tested in the same manner as sur- 
vival rates. Data from South Dakota indicated that males 
had a higher recovery rate than females (1.9 vs. 1.4%: 
P < 0.01). No differences were detected in each of the 
other four areas tested. All Z values were of the same sign, 
in favor of males having higher recovery rates, and the com- 
posite test statistic indicated males had a higher recovery 
rate than did females (P < 0.01). 
A nonparametric matched-pairs signed-ranks test (Siegel 
1956) was also used to test the hypothesis that direct 
recovery rates (see Glossary for differences between Re- 
covery Rates and Direct Recovery Rates) of male mourn- 
ing doves are similar to those of females. This test utilized 
more of the available data than either Option 1 of BRowNIE 
or the Z-test since the latter two utilized data from only 
8 and 5 of 15 reference areas, respectively. Male and female 
direct recovery rates for each State and each year were used 
as the matched-pairs of data. Of the 110 usable matched- 
pairs, 63 (57% ) indicated a higher recovery rate for males 
and 47 (43%) for females. The test failed to indicate a dif- 
ference (P > 0.10), 
Composite evaluation, — Although testing results are con- 
tradictory concerning differential survival and recovery 
rates due to sex, there probably were no real differences, 
This conclusion relates to the entire population and not to 
specific areas. Caution is suggested when inferences are 
drawn from data sets which are more restricted in scope 
than those used here. 
Age Related Differences 
A nonparametric matched-pairs signed-ranks test (Siegel 
1956) was used to test the hypothesis that survival rates of 
adult mourning doves (Table A-73) were similar to those 
of immatures (Table A-74). The test indicated that the sur- 
vival rates of adults and immatures differed (P < 0.05). 
Of the 11 areas used in the test, adults survived at a higher 
rate than immatures in all but Nebraska. 
A similar test was used to compare recovery rates of adult 
and immature doves. Rather than using the estimated 
annual recovery rates provided by program BRowNIE, actual 
direct recovery rates for each year and reference area were 
used (data obtained from printout not provided in this 
report). This increased the number of paired observations 
from 11 to 117 and increased the power of the test. Of the 
117 usable paired observations, 85 (73% ) indicated a higher 
direct recovery rate for immatures and 32 (27%) indicated 
a higher rate for adults. The composite difference was sig- 
nificant (P < 0.01) and we conclude that immatures were 
recovered at a higher rate than adults. 
Geographical Differences 
Among Central Management Unit States. — Both the esti- 
mated annual recovery rates (Tables A-73 and A-74) and 
direct recovery rates (Table A-18) varied among reference 
