Details
| Original language | English |
|---|---|
| Pages (from-to) | 29-35 |
| Number of pages | 7 |
| Journal | Journal of Agricultural, Biological, and Environmental Statistics |
| Volume | 8 |
| Issue number | 1 |
| Publication status | Published - Mar 2003 |
Abstract
The aquatic warbler, Acrocephalus paludicola, is one of a few species in which nestlings from a single nest can be sired by up to four different fathers. Data from a DNA fingerprinting study suggest that there is a trend: the larger the number of fathers in a brood the larger is the number of nestlings. However, the number of young within a single nest cannot be smaller than the number of fathers. This restriction causes an inherent trend and, consequently, can lead to a false-positive trend test result. For this nonstandard situation, we propose a trend test that differentiates between the inherent trend and a "real" trend, that is, a larger brood size through multiple paternity. Using DNA fingerprinting data, we performed a randomization test using the proposed new trend test statistic and obtained a significant result (P = 0.047). This indicates that a larger number of fathers per brood is associated with a larger brood size. In addition, we consider an umbrella alternative, that is, a downturn in effect may occur after the optimal number of fathers per brood is exceeded. An appropriate test for this alternative also leads to a significance (P = 0.011) .
Keywords
- DNA fingerprinting, Jonckheere-Terpstra test, Nonstandard situation, Randomization test, Umbrella alternative
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
- Agricultural and Biological Sciences(all)
- Agricultural and Biological Sciences (miscellaneous)
- Environmental Science(all)
- General Environmental Science
- Agricultural and Biological Sciences(all)
- General Agricultural and Biological Sciences
- Decision Sciences(all)
- Statistics, Probability and Uncertainty
- Mathematics(all)
- Applied Mathematics
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Journal of Agricultural, Biological, and Environmental Statistics, Vol. 8, No. 1, 03.2003, p. 29-35.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - A trend test for the analysis of multiple paternity
AU - Neuhäuser, Markus
AU - Leisler, Bernd
AU - Hothorn, Ludwig A.
PY - 2003/3
Y1 - 2003/3
N2 - The aquatic warbler, Acrocephalus paludicola, is one of a few species in which nestlings from a single nest can be sired by up to four different fathers. Data from a DNA fingerprinting study suggest that there is a trend: the larger the number of fathers in a brood the larger is the number of nestlings. However, the number of young within a single nest cannot be smaller than the number of fathers. This restriction causes an inherent trend and, consequently, can lead to a false-positive trend test result. For this nonstandard situation, we propose a trend test that differentiates between the inherent trend and a "real" trend, that is, a larger brood size through multiple paternity. Using DNA fingerprinting data, we performed a randomization test using the proposed new trend test statistic and obtained a significant result (P = 0.047). This indicates that a larger number of fathers per brood is associated with a larger brood size. In addition, we consider an umbrella alternative, that is, a downturn in effect may occur after the optimal number of fathers per brood is exceeded. An appropriate test for this alternative also leads to a significance (P = 0.011) .
AB - The aquatic warbler, Acrocephalus paludicola, is one of a few species in which nestlings from a single nest can be sired by up to four different fathers. Data from a DNA fingerprinting study suggest that there is a trend: the larger the number of fathers in a brood the larger is the number of nestlings. However, the number of young within a single nest cannot be smaller than the number of fathers. This restriction causes an inherent trend and, consequently, can lead to a false-positive trend test result. For this nonstandard situation, we propose a trend test that differentiates between the inherent trend and a "real" trend, that is, a larger brood size through multiple paternity. Using DNA fingerprinting data, we performed a randomization test using the proposed new trend test statistic and obtained a significant result (P = 0.047). This indicates that a larger number of fathers per brood is associated with a larger brood size. In addition, we consider an umbrella alternative, that is, a downturn in effect may occur after the optimal number of fathers per brood is exceeded. An appropriate test for this alternative also leads to a significance (P = 0.011) .
KW - DNA fingerprinting
KW - Jonckheere-Terpstra test
KW - Nonstandard situation
KW - Randomization test
KW - Umbrella alternative
UR - http://www.scopus.com/inward/record.url?scp=5444274111&partnerID=8YFLogxK
U2 - 10.1198/1085711031184
DO - 10.1198/1085711031184
M3 - Article
AN - SCOPUS:5444274111
VL - 8
SP - 29
EP - 35
JO - Journal of Agricultural, Biological, and Environmental Statistics
JF - Journal of Agricultural, Biological, and Environmental Statistics
SN - 1085-7117
IS - 1
ER -