Maureen Tingley

Small-Sample Confidence Intervals

In this article we present a technique for constructing oneor two-sided confidence intervals, which are second-order correct in terms of coverage, for either parametric or nonparametric problems. The construction is valid in the presence of nuisance parameters. The situation we consider is this: there are p parameters and we want a confidence interval for some function of them, possibly one of the parameters itself. The p parameters are estimated by M-estimates, which means they are obtained as a solution of a system of equations. Maximum likelihood estimates are included as a special case. The essential intermediate result, given in Equation (3.4), says that the estimated parameter of interest, 0, can be written as a mean, up to order op(l/ V/ ). The representation (3.4) is attained when 0 = 0(a) is a smooth function of the parameters q and aj is the solution of a well-behaved system of equations. We avoid the use of pivots and strive to obtain accurate coverage. Confidence intervals are constructed from a series of tests for the natural parameter of a one-parameter exponential family. We use the Lugannani and Rice (1980) tail area approximation to calculate bootstrap P values for the test statistic.

Bilateral isokinetic training reduces the bilateral leg strength deficit for both old and young adults

The bilateral limb deficit (BLD) describes the difference in maximal or near-maximal force generating capacity of muscles when they are contracted alone and in combination with the contralateral muscles. This study examined the effects of a 6-week (three times per week) bilateral leg strength training programme on BLD in younger and older adults. Data were collected from 33 subjects during slow (45°/s) isokinetic knee extensions and flexions before and after the training programme. After training, the BLD was reduced for extension (73.3–86.9%; P<0.001) but not for flexion (67.5–71.2%; P=0.13) regardless of age and gender. This study suggests that difficulty in recruiting all muscle units during a task involving bilateral activation can be improved by training, although such an effect appears to depend on the muscle group appreciated.

Using perceptual syntax to enhance semantic content in diagrams

Diagrams are essential in documenting large information systems. They capture, communicate, and leverage knowledge indispensable for solving problems and act as cognitive externalizations (intertwining internal and external processes to extract information from the external world to enhance thought). A diagram provides a mapping from the problem domain to the visual representation by supporting cognitive processes that involve perceptual pattern finding and cognitive symbolic operations. However, not all mappings are equal, and for effectiveness we must embed a diagrams representation with characteristics, which lets users easily perceive meaningful patterns. Consequently, a diagrams effectiveness depends to some extent on how well we construct it as an input to our visual system. In our research, we focus on a class of diagrams commonly referred to as graphs or node-link diagrams. Nodes representing entities, objects, or processes, and links or edges representing relationships between the nodes characterize them. Their most common form is outline circles or boxes denoting nodes and lines of different types representing links between the nodes. Entity-relationship diagrams, software structure diagrams, and data-flow models are examples of node-link diagrams used to model the structure of processes, software, or data.

An index to quantify normality of gait in young children.

Gait patterns are often described by recording the changes in angular rotation of such joints as the hip, knee and ankle, during a complete cycle. Each joint exhibits distinctive behavior throughout the gait cycle, and abnormal gait can be described by measuring departure from a typical (mean) joint rotation curve. Standard techniques for observation of gait patterns produce large sets of data. Data reduction is achieved in this work by locating primary directions of variation from mean behavior. Variation from the mean can then be summarized with a one-dimensional statistic, thought of as a squared distance from the population mean. Percentiles of this one-dimensional index can be calculated, enabling classification of a child as normal, unusual or abnormal. A key feature of this analysis is that it is applied across multiple joint angle curves and their derivatives, thus providing a measure that takes account of the interactions between the curves as well as their individual characteristics. A data base of 348 gait cycles, collected from normal children, aged 3-7, were analyzed. Data on each child were stored in a 36-dimensional vector. Most information on patterns of variation among normal children can be stored in a smaller 11-dimensional vector, which can be used for diagnostic purposes. Performance of the one-dimensional index of gait is demonstrated on data from very young children, and on children, up to age 7, who were born prematurely.

Effects of pulp and paper mill effluent on fish: a temporal assessment of fish health across sampling cycles.

The Canadian environmental effects monitoring (EEM) program is a regulated, cyclical, industry-funded program designed to determine whether receiving water impacts exist when a mill is in compliance with its discharge limits. The results from three cycles of the fish monitoring program (1992 to 2004) are available from over 200 surveys of fish compared between sites located upstream and downstream of pulp and paper mill effluent outfalls. Previous meta-analyses have shown a national average response pattern across cycles characterized by an increase in endpoints measuring energy storage and growth and a decrease in a reproductive endpoint, consistent with a response of nutrient enrichment in combination with some form of metabolic disruption. Although the national average pattern of effects was temporally consistent, there was some variability in the magnitude of effects among cycles. Questions were raised as to whether the intercycle variability was due to changes in effluent quality or due, at least in part, to other factors. The present study compares responses over the first three cycles, and shows that the choice of sentinel species is likely to be a major contributing factor to the variability in observed effects. Subset analyses using studies from mills that used the same sentinel species across cycles reveal fairly uniform responses and little evidence of significant improvements in overall fish health from cycles one to three. However, a meta-analysis using 1991 data collected from 10 mills before the implementation of the EEM program and data from the same mills collected during cycles one to three of the program reveal significantly reduced effects on relative liver weight and potential improvements in other endpoints.

Effect of stiffness and movement speed on selected dynamic torque characteristics of hydraulic-actuation joystick controls for heavy vehicles

The purpose of this work was to quantify the effects of joystick stiffness and movement speed on the dynamic torque characteristics of hydraulic-actuation joystick controls, as found in off-road vehicles, as one of the initial steps towards the development of a joystick design protocol. Using a previously developed mathematical model in which a hydraulic-actuation joystick is assumed to rotate about two axes where the rotation origin is a universal joint, the dynamic torque characteristics incurred by an operator were predicted. Utilizing a laboratory mock-up of an excavator cab environment, three actuation torque characteristics (peak torque, angular impulse and deceleration at the hard endpoint) were quantified for nine unskilled joystick operators during the use of a commonly used North American hydraulic-actuation joystick. The six different experimental conditions included combinations of three joystick stiffnesses and two movement speeds. The highest instantaneous input torque over the course of the joystick movement (not including the hard endpoint) was evaluated using the peak torque value. Angular impulse provided an indication of the sustained exposure to force. The third indicator, deceleration at the hard endpoint, was included to provide a description of impact loading on the hand as the joystick came to a sudden stop. The most important result of this work is that the dynamic torque characteristics incurred during hydraulic-actuation joystick use are substantial. While the peak torque values were not very different between the fast and slow motion conditions, the high decelerations even for slow movements observed at maximum excursion of the joystick indicate that the dynamics do matter. On the basis of deceleration at the hard endpoint and peak torque, the joystick movements that require the highest values for a combination of torque variables are the side-to-side ones. This suggests that less stiff balance and return springs should be considered for these directions than for forward and backward movements. However, if the design does not minimize acceleration, it is important that the spring stiffness not be too low since deceleration at the joystick hard endpoint will be very high causing the operator to incur large palm and finger impacts.

Dealing with heterogeneous regression slopes in analysis of covariance: new methodology applied to environmental effects monitoring fish survey data

Analysis of covariance (ANCOVA) is a powerful statistical method which incorporates one or more covariates into the analysis to reduce error associated with measurement. ANCOVA (modeling response as a function of fish size) is frequently used to analyze environmental effects monitoring (EEM) fish survey data. In approximately 12% of fish survey data sets taken from cycles 1 to 3 of Environment Canada’s EEM database for pulp and paper mills, the standard assumption of parallel regression slopes is not met. For the first three cycles of the EEM program, these data sets were classified as indicating a mill effect, but for the most part were excluded from subsequent analyses aimed at quantifying the effect. We present two different methods for initially dealing with data sets that exhibit heterogeneous slopes so that they can be analyzed using the parallel slope model. The first method identifies data sets where heterogeneous slopes are forced by a few high-influence observations. The second approach identifies data sets where a model with heterogeneous slopes is statistically, but not practically, significant: with a high coefficient of determination for the parallel slope model. These new methodologies are applied to EEM pulp and paper data sets and about 55% of cases with heterogeneous slopes can be described by a parallel slope model. We also discuss a third method that can be used to describe mill effects when regression slopes remain heterogeneous even after applying the above two methods, enabling comparison with a critical effect size. These new methodologies could benefit the EEM program by enabling more data sets to be incorporated into meta-analyses and be used to make more equitable mill monitoring decisions in the future.

Effect of joystick stiffness, movement speed and movement direction on joystick and upper limb kinematics when using hydraulic-actuation joystick controls in heavy vehicles

Despite the widespread use of hydraulic-actuation joysticks in mobile North American construction, mining and forestry vehicles, the biomechanical effects that joysticks have on their human operators has not been studied extensively. Using nine unskilled joystick operators and a laboratory mock-up with a commonly used North American heavy off-road equipment hydraulic-actuation joystick and operator seat, the purpose of this work was to quantify and compare the effects of three hydraulic-actuation joystick stiffnesses and two movement speeds on upper limb and joystick kinematics as one of the initial steps towards the development of a hydraulic-actuation joystick design protocol. In addition to providing a detailed description of the kinematics of a constrained occupational task, coupled with the corresponding effects of the task on operator upper limb kinematics, results from principal component analysis and ANOVA procedures revealed a number of differences in joystick and upper limb angle ranges and movement curve shapes resulting from the various joystick stiffness-speed combinations tested. For the most part, these joystick motion alterations were caused by small, insignificant changes in one or more upper limb joint angles. The two exceptions occurred for forward movements of the joystick; the fast speed – light stiffness condition movement pattern shape change was caused primarily by an alteration of the elbow flexion–extension movement pattern. Similarly, the fast speed – normal stiffness condition movement curve shape perturbation – was caused principally by a combination of significant movement curve shape alterations to elbow flexion–extension, external–internal shoulder rotation and flexion–extension of the shoulder. The finding that joystick stiffness and speed alterations affect joystick and upper limb kinematics minimally indicates that the joystick design approach of modelling the joystick and operator upper limb as a closed linkage system should be pursued. This approach would allow one to simulate the upper limb and joystick kinematics that result from virtual changes to upper limb and joystick lengths.

Comparison of two normative paediatric gait databases

The availability of age-matched normative data is an essential component of clinical gait analyses. Comparison of normative gait databases is difficult due to the high-dimensionality and temporal nature of the various gait waveforms. The purpose of this study was to provide a method of comparing the sagittal joint angle data between two normative databases. We compared a modern gait database to the historical San Diego database using statistical classifiers developed by Tingley et al. (2002). Gait data were recorded from 60 children aged 1–13 years. A six-camera Vicon 512 motion analysis system and two force plates were utilized to obtain temporal-spatial, kinematic, and kinetic parameters during walking. Differences between the two normative data sets were explored using the classifier index scores, and the mean and covariance structure of the joint angle data from each lab. Significant differences in sagittal angle data between the two databases were identified and attributed to technological advances and data processing techniques (data smoothing, sampling, and joint angle approximations). This work provides a simple method of database comparison using trainable statistical classifiers.

A Note on Obtaining Confidence Intervals for Discrete Parameters

Abstract Peskun considers the problem of constructing a closed confidence interval for a single parameter, using a discrete test statistic. The interval is constructed by generalizing the method of Konijn. Peskun gives an example where the interval fails to be well defined. In this article a confidence interval is constructed by generalizing the method of Cochran, in a manner similar to the way in which Peskun generalizes Konijns technique. Buonaccorsi compares the Konijn and Cochran type intervals in the hypergeometric setting. This article shows that the Cochran interval is always well defined and contains the Peskun interval, and that the Cochran interval contains, at most, two points more than the Peskun interval. Discrepancies between the two kinds of intervals, and ill definition of the Peskun interval, follow from discreteness of the parameter space, rather than discreteness of the test statistic.