In this paper, we will use Cochran-Mantel-Haenszel (CMH) statistics to analyse four data sets which have appeared in the literature. It is well known that tests based on the CMH statistics are equivalent to certain standard rank tests but here we show how CMH statistics also apply in less standard situations. In particular, examples are given for randomized block designs both with and without m...

Key distribution is an important problem in wireless sensor networks where sensor nodesare randomly scattered in adversarial environments.Due to the random deployment of sensors, a list of keys must be pre-distributed to each sensor node before deployment. To establish a secure communication, two nodes must share common key from their key-rings. Otherwise, they can find a key- path in which ens...

The existence of pairwise additive balanced incomplete block (BIB) designs and pairwise additive cyclic BIB designs with λ = 1 has been discussed through direct and recursive constructions in the literature. This paper takes BIB designs with 1-rotational automorphisms and then the existence of pairwise additive 1-rotational BIB designs is investigated for λ = 1. Finally, it is shown that there ...

A double-blind, crossover clinical trial was carried out in subjects with fever, to assess the activity of three antipyretic drugs. Eighteen in-patients with body temperature (axillary) above 38 degrees C were given a single dose of nimesulide 100 mg, acetylsalicylic acid 500 mg, and dipyrone 500 mg, orally according to an incomplete balanced block design. Significant improvement was recorded w...

This research examined the impact of the length and format (video, audio or text) of instructional Web modules (segments of lectures) on information recall, persistence, perceived content difficulty, aesthetic appeal, and perceived module length. Twelve Web-based instructional modules with different formats and lengths were evaluated through an experiment with a balanced-incomplete block design...

In this paper we propose a method for scaling up filterbased feature selection in classification problems. We use the conditional mutual information as filter measure and show how the required statistics can be computed in parallel avoiding unnecessary calculations. The distribution of the calculations between the available computing units is determined based on balanced incomplete block design...

Conditions under which constant block size implies constant replication number and vice versa are established for balanced ternary designs. This note is motivated by a note of almost the identical title of W. D. Wallis [2]. In a balanced binary design with constant block size, the replication number must be constant, but the converse is not true. Wallis [2] proved the following partial converse...