A Test for Symmetry in Contingency Tables

Bowker, Albert H.

doi:10.2307/2280710

Cited by 203 publications

(156 citation statements)

References 0 publications

Supporting

Mentioning

153

Contrasting

Unclassified

Order By: Relevance

“…Quality control was maintained by the readers recounting earlier samples prior to initiating each reading and crosschecking the readings. The consistency of age estimates between ageing rounds and readers was assessed using age bias plots and chi-square tests of symmetry (McNemar 1947;Bowker 1948;Campana et al 1995;Hoenig et al 1995;Evans and Hoenig 1998). To resolve discrepancies between readers greater than 2 years, a consensus was reached by reading the centra together.…”

Section: Age Determinationmentioning

confidence: 99%

Validated age and growth estimates for Carcharhinus obscurus in the northwestern Atlantic Ocean, with pre- and post management growth comparisons

Natanson

Gervelis

Winton³

et al. 2013

Environ Biol Fish

View full text Add to dashboard Cite

Age and growth estimates for the dusky shark, Carcharhinus obscurus, were derived from vertebral centra collected in the northwestern Atlantic Ocean. Sample collection spanned the years prior to and following the implementation of management measures . Growth was compared pre-and post-population depletion and pre-and post-management to investigate the possibility of density-mediated shifts in age and growth parameters over time. There was no evidence of difference between periods for either sex. Additionally, bomb radiocarbon dating was used to determine the periodicity of band pair formation. Results support the traditional interpretation of annual band pairs up to approximately 11 years of age. After this time, vertebral counts considerably underestimate true age. Maximum validated ages were estimated to be between 38 and 42 years of age (an increase of 15 to 19 years over the band count estimates), confirming longevity to at least 42 years of age. Growth curves estimated using only validated data were compared to those generated using band pair counts. Logistic growth parameters derived from validated vertebral length-at-age data were L ∞ = 261.5 cm FL, L o =85.5 cm, t o =4.89 year and g= 0.15 year −1 for the sexes combined. Revised estimates of age at maturity were 17.4 years for males and 17.6 years for females.

show abstract

Section: Age Determinationmentioning

confidence: 99%

Validated age and growth estimates for Carcharhinus obscurus in the northwestern Atlantic Ocean, with pre- and post management growth comparisons

Natanson

Gervelis

Winton³

et al. 2013

Environ Biol Fish

View full text Add to dashboard Cite

show abstract

“…Taking the benchmarks proposed by Landis and Koch (1977) as a baseline this can be classified as a fair agreement and association between choices (kappa statistic of 0.38). The test proposed by Bowker (1948), suitable for dependent samples, does not allow a rejection of the null hypothesis of symmetry between test and retest at the 5 %-significance level with regard to the total of five choice sets (Chi-square = 3.05, 3 df, p = 0.384).…”

Section: Test-retest Reliability At the Level Of Choice Sets And The mentioning

confidence: 99%

Test–Retest Reliability of Choice Experiments in Environmental Valuation

Liebe

Meyerhoff

Hartje

2012

Environ Resource Econ

View full text Add to dashboard Cite

The paper presents the results of the first test-retest study on choice experiments in environmental valuation. In a survey concerning landscape externalities of onshore wind power in central Germany, respondents answered the same five choice sets at two different points in time. Each choice set comprised three alternatives described by five attributes, and the time interval between the test and the retest was eleven months. The analysis takes place at three different levels, investigating choice consistency at the choice task level and repeatability of the latent construct utility at the level of parametric models as well as at the level of willingness-to-pay estimates. At the choice task level we observed 59 % identical choices. The parametric analysis shows that the test and retest estimates are not equal, even when we control for scale, that is, differences in the error variance. However, comparing the marginal willingness-to-pay estimates among test and retest reveals only a statistically significant difference for one of the attributes. Overall, this indicates a moderate test-retest reliability taking into account that consistency at the choice task level overlooks the stochastic nature of the process underlying discrete choice experiments.

show abstract

“…Pairs of scores obtained from the consensus and independent methods were compared within patients and the weighted kappa statistic calculated as a measure of agreement. The test of symmetry [16] was used to test whether one method produced higher scores for change in breast appearance compared with the other method of assessment.…”

Section: Methodsmentioning

confidence: 99%

Evaluation of a Method for Grading Late Photographic Change in Breast Appearance after Radiotherapy for Early Breast Cancer

Haviland

Ashton²,

Broad³

et al. 2008

Clinical Oncology

View full text Add to dashboard Cite

ABSTRACT:Aims: Serial photographs have been collected prospectively to evaluate the effect of radiotherapy on normal tissues in the breast. The aim of this study was to compare two methods of scoring radiation-induced changes. Materials and methods: Five-year photographs of 400 patients randomised to receive either 42.9 or 39 Gy in 13 fractions to the whole breast after tumour excision of early breast cancer were compared with a post-surgery baseline and scored for change in breast appearance on a three-point graded scale. Two alternative methods of scoring using three observers were compared: (a) scores allocated independently, with independent resolution of discrepancies, and (b) scores allocated by consensus. Results: Treatment effects estimated from the consensus and independent scores were very similar (odds ratio 1.89, 95% confidence interval 1.21e2.96 vs 2.28, 95% confidence interval 1.50e3.47, respectively). Agreement between the scores obtained from each method was reasonable, and the repeatability of the consensus method was good. Conclusions: The consensus method of scoring photographic change in breast appearance seems to be no less sensitive to randomised dose as the independent method of assessment, but is much quicker to administer. The consensus method has been used to score over 3000 sets of photographs in the National Cancer Research Institute Standardisation of Breast

show abstract

A Test for Symmetry in Contingency Tables

Cited by 203 publications

References 0 publications

Validated age and growth estimates for Carcharhinus obscurus in the northwestern Atlantic Ocean, with pre- and post management growth comparisons

Validated age and growth estimates for Carcharhinus obscurus in the northwestern Atlantic Ocean, with pre- and post management growth comparisons

Test–Retest Reliability of Choice Experiments in Environmental Valuation

Evaluation of a Method for Grading Late Photographic Change in Breast Appearance after Radiotherapy for Early Breast Cancer

Contact Info

Product

Resources

About