JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Biometrika Trust is collaborating with JSTOR to digitize, preserve and extend access to Biometrika. SUMMARY Two C(a) statistics (Neyman, 1959) for testing the equality of the means of several groups of count data in the presence of a common dispersion parameter are derived. The performance of the two C(a) statistics, the likelihood ratio statistic and two more statistics based on transformed data (Anscombe, 1948) is then compared in terms of size and power by using Monte Carlo simulations. The C(a) statistics are recommended. The C (a) statistics for testing the equality of the dispersion parameters of several groups of count data are also derived. Two applications are given. Some key words: Common dispersion parameter; C(a) statistic; Equality of means; Equality of dispersion parameters; Negative binomial distribution.

We obtain a single equation for the maximum likelihood estimator of a common intraclass correlation ρ based on L ≥ 2 independent samples from multivariate normal populations. This equation, which is to be solved iteratively, is similar in form to that obtained by Pearson (1933) for the estimation of a common correlation from several bivariate normal populations. Further, we derive a C(α) statistic (Neyman, 1959) for testing the equality of several intraclass correlations. This statistic is very simple in form and is expected to maintain nominal level well, at least asymptotically. Two examples are given.

No abstract

This paper is concerned with the detection of a single (upper or lower) outlier in a Poisson sample. A conditional test due to Doornbos (1966) is discussed and some unconditional tests based on the original and transformed data are developed. Of the unconditional tests the critical values of the likelihood ratio statistic LRT and the statistic M based on adjusted standardized residuals show dependence on the Poisson Parameter. The Dixon-type statistics based on transformed sample produce critical values which are nearly parameter-independent and in reasonably close agreement with those from the normal distribution. For reasons of simplicity and power we recommend using the * statistic T(T ) for detecting an upper (lower) outlier, where ' i and Xi Poisson ( A ) . This recommendation = Jxi + 6 falls in line with Barnett and Lewis (1978).