Search

Researchers typically assume that they are working from a normal distribution and with independent sampling. Both assumptions are often violated. Our goal was to explore the intersection of the violations: Is the net effect good or bad? Using the family of skew-normal distributions, which is a superset of the family of normal distributions, we tested whether the mean squared error (MSE) is less under dependence or under independence. We found that the MSE is less under dependence, under the assumption that elements in both samples are identically distributed related to the population distribution. In addition, increasing skewness and increasing sample size also decrease the MSE. Finally, the largest differences in MSE between dependence and independence occur under moderate skewness. [ABSTRACT FROM AUTHOR]

Previous researchers have proposed the a priori procedure, whereby the researcher specifies, prior to data collection, how closely she wishes the sample means to approach corresponding population means, and the degree of confidence of meeting the specification. However, an important limitation of previous research is that researchers sometimes are interested in differences between means, rather than in the means themselves. To address this limitation, we propose additional equations that expand the a priori procedure to handle differences between means, both in matched and in independent samples. Finally, implications are discussed. [ABSTRACT FROM AUTHOR]

Standard approaches for comparing the means of two samples, comprising both paired observations and independent observations, involve the discarding of valuable information. An alternative test which uses all of the available data, is the partially overlapping samples t-test. Two variations of the test are available, one assuming equal variances, and one assuming separate variances. Issues with standard procedures, and considerations for choosing appropriate tests in the partially overlapping scenario are discussed. An example with details of how to apply the partially overlapping samples t-test is given along with an R package that implement these new tests.

The concept of quality management in the public sector is nowadays more popular than ever before. Since it forms in the fields of public interest, public companies, institutions and organizations are crucially important for the development and the functioning of any social and modern society. Therefore, those organisations are expected to perform transparent, efficient, democratic and socially righteous work with the goal of maintaining and improving the quality of a nation’s economic life. In Bosnia and Herzegovina, the public sector is specific and additionally complicated by a complex political system, legal procedures and the constitutional structure and as such, it is often very ineffective. The main purpose of the implementation of ISO 9001 is to introduce a system of quality management and improve the quality of public services as well as to ensure customer satisfaction. Therefore, the goal of this paper is to present the concept of customer satisfaction and selected statistical methods in quality management, and to apply a part of these methods through secondary research and analysis of published reports about customer satisfaction with services of the public sector in Bosnia and Herzegovina. Results of the study have confirmed greater user satisfaction with service in the public sector in units of local self-government with an implemented ISO 9001 standard implemented ISO 9001 standard than in units in which the standard has not been implemented.

Markov chains (MCs) are convenient means of generating realizations of networks with a given (joint or otherwise) degree distribution (DD), since they simply require a procedure for rewiring edges. The major challenge is to find the right number of steps to run such a chain, so that we generate truly independent samples. Theoretical bounds for mixing times of these MCs are too large to be practically useful. Practitioners have no useful guide for choosing the length, and tend to pick numbers fairly arbitrarily.We give a principled mathematical argument showing that it suffices for the length to be proportional to the number of the desired number of edges. We also prescribe a method for choosing this proportionality constant. We run a series of experiments showing that the distributions of common graph properties converge in this time, providing empirical evidence for our claims. [ABSTRACT FROM AUTHOR]

Prediction of censored order statistics from a Type-II censored sample can be done with trivial bounds having perfect confidence. However, given independent samples from the same absolutely continuous distribution, improved bounds can be attained. In this regard, we develop here point prediction based on L-statistics for predicting order statistics (OS) from a future sample as well as for predicting censored OS from a Type-II censored sample. An example is taken to illustrate these ideas, and the limiting case wherein a single independent sample is arbitrarily large is also discussed. [ABSTRACT FROM AUTHOR]

The turn-table platform stirring technique is used to improve the measurement accuracy of a reverberation chamber (RC). It is usually assumed that the half-wavelength platform spatial sampling results in independent samples, which can be used to approximate the independent platform sample number. But the approximation is rather coarse. This paper presents a measurement-based method to determine the independent platform sample number. This method is simple and accurate. In contrast to the half-wavelength coarse approximation that depends only on frequency, this method in addition illustrates the dependence of the independent platform sample number on RC loading and mode stirring. [ABSTRACT FROM AUTHOR]

We derive equations to calculate the number of samples (tuner positions) required to achieve a certain inhomogeneity for the maximum of a rectangular field strength component with given confidence in reverberation chambers. Perfect and limited reverberation are investigated. The results are in good agreement with unreferenced values found in reverberation standards, e.g., IEC 61000-4-21. They can be used to a priori predict the required number of independent samples from the quality factor and the chamber volume only. Evidence is given that current homogeneity limits in standards (3 dB, IEC 61000-4-21) are inconsistent with the minimum number of samples to use (namely 12) at high frequencies. [ABSTRACT FROM AUTHOR]

The aim of this study was to investigate whether differences in information and communication technology (ICT) readiness and access across countries were fundamentally related to the variable of English language adoption. A one-way multivariate analysis of variance (MANOVA) was utilized to comparatively examine the developments of ICT readiness and use among 149 countries categorized into four groups based on levels of English language adoption and economic development. The results of the comparative analysis showed that English language adoption is not a dominant factor in determining the global digital divide. The results suggest that much of the differences in ICT development across countries are attributable to levels of economic development. Limitations and implications for additional studies on specifying key factors widening the global technology gap are discussed. [ABSTRACT FROM AUTHOR]

In reverberation chambers (RCs), measurements are usually performed by changing the boundary conditions using a mode stirrer. The major difficulty is to select Uncorrelated samples in order to make a statistical analysis of the data. Furthermore, the knowledge of the number of independent samples is of crucial importance to assess the measurement accuracy. To evaluate whether measured data are independent, the conventional method compares the autocorrelation function (ACF) with the critical value 0.37. However, this criterion is generally not appropriate because the ACF probability density function (pdf) depends strongly on the sample size. For a measurement series of length N, the effective sample size (ESS) is defined as the number N′ < N of independent samples, which would provide the same information as the N-size sample. This paper aims to provide a new method based on autoregressive (AR) models and the central limit theorem (CLT) in the case of dependent data, for estimating the ESS. The proposed method is easy to implement since it requires only the knowledge of simple statistical parameters. Moreover, it provides useful guidelines to assess the maximum number of independent samples available with the mode stirrer. Experimental results are in good agreement with the theoretical models, either for the electric field or the received power. [ABSTRACT FROM AUTHOR]

This article focuses on a new method for producing estimates from a multiple frame survey. This method was developed in connection with a Statistics Canada survey of unincorporated businesses based on tax data. An initial stratified simple random sample is selected from a tax data frame that is incomplete and whose data are at least one year old. Stratification by type of industry, however, is possible. A subsequent simple random sample stratified on gross business income is selected from a second tax data frame that is complete and current. Stratification by type of industry, however, is not possible in this subsequent sample. The tax forms selected in either of these two independent samples are located and the data required for statistical purposes are transcribed. The sampled data are then weighted to produce estimates based on both samples. In a two-frame sample design, usually Frame A and Frame B each contains units the other frame does not have as well as units in common. Selecting a stratified sample from Frame A can be viewed as equivalent to selecting a stratified sample from the complete population in which one stratum has no sample selected from it.

Explanations of advantages and disadvantages of paired-samples experimental designs in textbooks in education and psychology frequently overlook the change in the Type I error probability which occurs when an independent-samples t test is performed on correlated observations. This alteration of the significance level can be extreme even if the correlation is small. By comparison, the loss of power of the paired-samples t test on difference scores due to reduction of degrees of freedom, which typically is emphasized, is relatively slight. Although paired-samples designs are appropriate and widely used when there is a natural correspondence or pairing of scores, researchers have not often considered the implications of undetected correlation between supposedly independent samples in the absence of explicit pairing. [ABSTRACT FROM PUBLISHER]

In quantitative research, the selection of the most appropriate statistical test for the comparison of two independent samples can be problematic. There is a lack of consensus in the statistics community regarding the appropriate approach, particularly towards assessing assumptions of normality and equal variances. The lack of clarity in the appropriate strategy affects the reproducibility of results. Statistical packages performing different tests under the same name only adds to this issue. The process of preliminary testing assumptions of a test using the sample data, before performing a test conditional upon the outcome of the preliminary test, is performed by some researchers; this practice is often criticised in the literature. Preliminary testing is typically performed at the arbitrary 5% significance level. In this paper, this process is reviewed and additional results are given using simulation, examining a procedure with normality and equal variance preliminary tests to compare two independent samples. [ABSTRACT FROM AUTHOR]