Sampling distributions for introductory statistics students using internet polling software

The topic of sampling distributions is the first source of considerable difficulty for introductory statistics students. In this article, an active learning idea for an introduction to sampling distributions is presented. The activity uses real data collected from the class, and the results are used to generate sampling distributions using electronic communication of sample statistics from student mobile telephones.     

A first course in biostatistics for health sciences students

The content of a course on introductory biostatistics for health science students is described. The course, first taught in 1998, embodies, among other topics, most of the desirable features discussed by Sahai and Ojeda. The syllabus and associated project work are presented with emphasis placed on ways in which study design and critical evaluation of research are developed in parallel with the statistical methodology. The advantages and disadvantages of such a course are discussed. This includes an analysis of the performance of students with diff erent statistical and mathematical backgrounds.     

Attitudes and perceived usefulness of statistics among health sciences students

A questionnaire was completed at the cessation of semester 2 in November 1989 by 102 postgraduate and 58 undergraduate health sciences students studying introductory statistics units. The questionnaire measured; (i) the student's attitude towards statistics, (ii) the way in which they learned statistics, (iii) the student's intention to pursue further statistics training, and (iv) the perceived usefulness of statistics in their professions. It was found that the learning of statistics would be enhanced by smaller tutorial groups, and more exposure to computer printouts to assist in interpretation of results. An emphasis on the understanding of statistics presented in journal articles should be a priority. It was also apparent that the more computer and research experience the student had prior to commencing the course, the greater the likelihood of a positive attitude towards statistics. However, both undergraduate and postgraduate students indicated that they would not enrol in an advanced biostatistics course, but would rather consult a statistician when necessary. Suggestions for more effective statistical teaching for health sciences students are also discussed.     

An invariance principle for linear combinations of order statistics

Summary Based on an (almost sure) reverse-martingale representation for linear combinations of order statistics (with smooth weight functions), a backward invariance principle (relating to the tail sequence) is established and the underlying regularity conditions are critically examined.Work supported by the Air Force Office of Scientific Research, AFSC, USAF, Contract No. AFOSR 74-2736. Reproduction in whole or part is permitted for any purpose of the U.S. Government     

Developing statistical modelers and thinkers in an introductory, tertiary-level statistics course

While models are an important concept in statistics, few introductory statistics courses at the tertiary level put models at the core of the curriculum. This paper reports on a radically different approach to teaching statistics at the tertiary level, one that uses models and simulation as the organizing theme of the course. The focus on modeling and simulation??along with inference??was facilitated by having students use TinkerPlots? software for all modeling and analysis. Results from a 3-month teaching experiment suggest that a course focused on modeling and simulation through randomization and resampling methods in which students learn to think using a powerful and conceptual modeling tool can foster ways of thinking statistically. Furthermore, such an approach seems to help students develop experiences with and appreciation for the science and practice of statistics.     

Identification and definition of lexically ambiguous words in statistics by tutors and students

Lexical ambiguity arises when a word from everyday English is used differently in a particular discipline, such as statistics. This paper reports on a project that begins by identifying tutors’ perceptions of words that are potentially lexically ambiguous to students, in two different ways. Students’ definitions of nine lexically ambiguous words are also collected at the beginning and end of a semester of introductory statistics study, in a complex design taking account of multiple tutors and multiple words in multiple contexts. Tutor perceptions and actual student difficulties at the beginning of a semester are compared. The lexical ambiguity associated with the word ‘significance’ is shown to be evident in students even after completing an introductory statistics course.     

An applied cell mapping method for optimal control problems

From the application point of view, a series of modifications are proposed for the cell mapping method discussed in Ref. 1 for the optimal control analysis of dynamical systems. The cell order around the target set is rearranged. A set of common discriminate principles is used for the selection of the optimal one among competing control strategies of the same cost. Inequality constraints of the system are taken into account. The number of elements in the set of allowable time intervals is not prescribed, but left open. These modifications seem to make the cell mapping method more efficient for analyzing feedback systems and for obtaining their global optimal control information. The algorithms presented in this paper could broaden the application of the cell mapping approach of Ref. 1 to a wider class of engineering problems.     

An evaluation of continuous assessment in statistics courses for social scientists

This paper reports on the use of continuous assessment in a statistics course for social scientists, in which assignments are open‐ended and use real data. The students' performance in the assignments and a formal examination are compared, and it is found that when students are assessed by a combination of continuous assessment and examination, those students who carry out the majority of the assignments score higher examination marks than do the rest.     

An extension of almost sure central limit theorem for order statistics

Let {X _n ,n ≥ 1} be a sequence of i.i.d. random variables. Let M _n and m _n denote the first and the second largest maxima. Assume that there are normalizing sequences a _n  > 0, b _n and a nondegenerate limit distribution G, such that . Assume also that {d _k ,k ≥ 1} are positive weights obeying some mild conditions. Then for x > y we have

Preemptive benchmarking problem: An approach for official statistics in small areas

National Statistical Agencies and Autonomous Institutions are extremely interested in using information from those areas that are actually smaller than the actual areas for which a survey is initially designed. As such, small area estimation and its application are valuable when conducting research on Official Statistics. A wide range of different methods are available which provide estimations to small area levels, being reasonable to guarantee that they add up to the published design-based estimations in a large area that includes these small areas. This requirement is known as benchmarking. Different algorithms, all based on distances between original data and modified data, are introduced in this paper, with the intention of satisfying the benchmarking property. We provide rules to apply these proposed calibrated methods according to user criteria. Goal programming with priorities methodology is used to represent user preferences. The result is a collection of different interdependent network flow problems. Some of these problems require the development of ad hoc methods. The introduced methods are assessed by a Monte Carlo simulation study using the Spanish Labour Force Survey in the Canary Islands. The results also show that the consistency of the estimator is independent of the used calibrated methods, but it does depend on the benchmarking weights.     

An efficient computational method for moments of order statistics under progressive censoring

Thomas and Wilson (Technometrics 14 (1972) 679) developed a computational method for calculating the single and product moments of order statistics from progressively censored samples by making use of the corresponding moments of the usual order statistics. The absence of an explicit representation for the marginal and joint density function of order statistics under progressive censoring makes their method extremely tedious. By deriving the required marginal and joint density functions in explicit form, we obtain an alternative, highly efficient, method for computing the desired moments.     

An integrated introduction to spreadsheet and programming skills for operational research students

Spreadsheets and computer programming make valuable contributions to OR courses. The latest version of Microsoft Excel, integrated with Visual Basic for Applications (VBA), supports a seamless transition from spreadsheets to programming for such courses. This paper describes a teaching approach for such a course. It is illustrated with examples that have been designed and developed to help students to make this transition, whilst supporting the operational research context of the course. It shows how VBA supports this transition as well as demonstrating its functional contribution to spreadsheet modelling. This paper goes on to discuss insights into spreadsheet design that are afforded by this software, in particular concerning differentiation of statics from dynamics, control, visibility of data, and user interaction. A further focus of discussion is the balance of functionality between worksheet structure and formulae, and VBA modules and functions. The course is well received by students, forms a significant contribution to the MSc programme that is its context, and offers a model to other similar programmes.     

An asymptotic expansion for product integration applied to Cauchy principal value integrals

Product integration methods for Cauchy principal value integrals based on piecewise Lagrangian interpolation are studied. It is shown that for this class of quadrature methods the truncation error has an asymptotic expansion in integer powers of the step-size, and that a method with an asymptotic expansion in even powers of the step-size does not exist. The relative merits of a quadrature method which employs values of both the integrand and its first derivative and for which the truncation error has an asymptotic expansion in even powers of the step-size are discussed.     

Kaizen teaching and the learning habits of engineering students in a freshman mathematics course

Which are the teaching methods that actually contribute to the learning of mathematics? The answer to this certainly is the holy grail of didactic and pedagogy, and should be supported by large scale statistical evidence. Our article aims at providing an initial step into this direction by first illustrating a teaching paradigm that is suited for the generation of large scale data sets: based on industry best practice quality assurance standards we introduce the Kaizen teaching paradigm which enforces Kolb’s reflective learning cycle on the students’ side. Second, we present and analyze the data we obtained through our pilot implementation at a engineering freshman mathematics course in the Sultanate of Oman. These emphasize the effectiveness of Kaizen teaching and once again show the necessity of continuous learning. A practice that seems to be forgotten in traditional university engineering courses due to the mere size of the audience. In particular it seems that a Markovian estimator for students’ performance may have to be considered.     

An evaluation of test statistics for detecting temporary change in Bl(1,1,1,1) models

A study is carried out to investigate the sampling properties of the outlier test statistics of a procedure developed for detecting temporary change in BL(1,1,1,1) processes. It is done with respect to the sample size, the type of outlier and the size of the coefficients of the BL(1,1,1,1) process. The results show that, in general, the outlier detection procedure is capable of detecting TC, although the performance is affected if ω is large. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Uniform estimates for order statistics and Orlicz functions

We establish uniform estimates for order statistics: Given a sequence of independent identically distributed random variables ξ ₁, … , ξ _n and a vector of scalars x = (x ₁, … , x _n ), and 1 ≤ k ≤ n, we provide estimates for \mathbb E   k-min_{1 ￡ i ￡ n} |x_ix_i|{\mathbb E \, \, k-{\rm min}_{1\leq i\leq n} |x_{i}\xi _{i}|} and \mathbb E k-max_{1 ￡ i ￡ n}|x_ix_i|{\mathbb E\,k-{\rm max}_{1\leq i\leq n}|x_{i}\xi_{i}|} in terms of the values k and the Orlicz norm ||y_x||_M{\|y_x\|_M} of the vector y _x  = (1/x ₁, … , 1/x _n ). Here M(t) is the appropriate Orlicz function associated with the distribution function of the random variable |ξ ₁|, G(t) = \mathbb P ({ |x₁| ￡ t}){G(t) =\mathbb P \left(\left\{ |\xi_1| \leq t\right\}\right)}. For example, if ξ ₁ is the standard N(0, 1) Gaussian random variable, then G(t) = ?{\tfrac2p}ò₀^t e^-\fracs22ds {G(t)= \sqrt{\tfrac{2}{\pi}}\int_{0}^t e^{-\frac{s^{2}}{2}}ds }  and M(s)=?{\tfrac2p}ò₀^se^-\frac12t2dt{M(s)=\sqrt{\tfrac{2}{\pi}}\int_{0}^{s}e^{-\frac{1}{2t^{2}}}dt}. We would like to emphasize that our estimates do not depend on the length n of the sequence.     

The Malthusian parameter in microbial ecology and evolution: An optimal control treatment

There is a consensus among evolutionary microbiologists that the (quasi-)instantaneous specific rate of proliferation is a good measure for fitness. On the other hand, it is well known in evolutionary theory that the appropriate fitness measure is generally dependent on ecotype. We consider two ecotypes: one in which the instantaneous specific growth rate, discounted into the future, is an appropriate measure of fitness, and one in which fitness is measured as “ burst fitness” under intermittently favorable conditions. Our analysis is based on a standard model of microbial growth, the Droop Quota model, which we have extended with a variable balance between assimilation and proliferation, since this balance constitutes an important regulatory choice. We evaluate optimal allocation regimes and contrast these to control regimes based on maximization of the instantaneous specific growth rate.