Showing papers on "Reliability (statistics) published in 1994"
Book•
30 Nov 1994
TL;DR: The basic concepts of Power System Reliability Evaluation and Elements of Monte Carlo Methods and Reliability Cost/Worth Assessment are explained.
Abstract: Introduction. Basic Concepts of Power System Reliability Evaluation. Elements of Monte Carlo Methods. Generating System Adequacy Assessment. Composite System Adequacy Assessment. Distribution System and Station Adequacy Assessment. Reliability Cost/Worth Assessment. Appendixes. Index.
1,459 citations
TL;DR: Video training and certification is a practical and effective method to standardize the use of examination scales in a study of acute stroke therapy so that all examiners rated patients comparably.
Abstract: Background and Purpose Despite the frequent use of clinical rating scales in multicenter therapeutic stroke trials, no generally acceptable method exists to train and certify investigators to use the instrument consistently. We desired to train investigators to use the National Institutes of Health Stroke Scale in a study of acute stroke therapy so that all examiners rated patients comparably. Methods We devised a two-camera videotape method that optimizes the visual presentation of examination findings. We then measured the effectiveness of the training by asking each investigator to evaluate a set of 11 patients, also on videotape. We tabulated the evaluations, devised a scoring system, and calculated measures of interobserver agreement among the participants in this study. Results We trained and certified 162 investigators. We found moderate to excellent agreement on most Stroke Scale
1,062 citations
Proceedings Article•
01 Jan 1994
TL;DR: This paper presents a reliability parameter, performance factor, to be used as a direct quantitative measure of customer satisfaction, and documents the roots of the performance factor metric, shows explicit examples of the use of theperformance factor, and presents a field example of thePerformance factor in a customer maintenance situation.
Abstract: Customer satisfaction is generally presented in quality terms. This paper presents a reliability parameter, performance factor, to be used as a direct quantitative measure of customer satisfaction. It documents the roots of the performance factor metric, shows explicit examples of the use of the performance factor, and presents a field example of the performance factor in a customer maintenance situation. From experience in commercial electronics, recommendations are made on making this reliability metric a realistic complement to quality as a measure of customer satisfaction. Another name for customer satisfaction is reliability. Fewer field failures mean fewer returns; fewer returns mean a larger field demonstrated MTBF, and higher actual MTBFs mean a higher performance factor. >
1,013 citations
TL;DR: The evidence supports the use of: (1) the 12-item instrument as an overall measure of EUCS; and (2) the five component factors for explaining the EUCS construct.
Abstract: The dimensionality of the user satisfaction construct is an important theoretical issue that has received considerable attention. Focusing on end users who directly interact with application software, Doll and Torkzadeah (1988) develop a 12-item application specific instrument for providing an overall assessment of end-user computing satisfaction (EUCS). They also contend that EUCS is comprised of five component measures (content, accuracy, format, ease of use, timeliness) that explain the construct and permit more precision in formulating and testing research hypotheses. The acceptance of the EUCS as a standardized instrument requires confirmation that it explains and measures the user satisfaction construct.Based on a sample of 409 respondents from 18 organizations, this research uses confirmatory factor analysis (LISREL) to test alternative models of underlying factor structure and assess the reliability and validity of factors and items. Without respecifying the model, the results provide some support for Doll and Torkzadeh's proposed model of the five first-order factors and one second-order (higher-order) factor. Thus the evidence supports the use of: (1) the 12-item instrument as an overall measure of EUCS; and (2) the five component factors for explaining the EUCS construct.
1,010 citations
Journal Article•
TL;DR: Validity and reliability are two important characteristics of behavioral measure and are referred to as credibility and reliability.
Abstract: For the statistical consultant working with social science researchers the estimation of reliability and validity is a task frequently encountered. Measurement issues differ in the social sciences in that they are related to the quantification of abstract, intangible and unobservable constructs. In many instances, then, the meaning of quantities is only inferred. Let us begin by a general description of the paradigm that we are dealing with. Most concepts in the behavioral sciences have meaning within the context of the theory that they are a part of. Each concept, thus, has an operational definition which is governed by the overarching theory. If a concept is involved in the testing of hypothesis to support the theory it has to be measured. So the first decision that the research is faced with is \" how shall the concept be measured? \" That is the type of measure. At a very broad level the type of measure can be observational, self-report, interview, etc. These types ultimately take shape of a more specific form like observation of ongoing activity, observing video-taped events, self-report measures like questionnaires that can be open-ended or close-ended, Likert-type scales, interviews that are structured, semi-structured or unstructured and open-ended or close-ended. Needless to say, each type of measure has specific types of issues that need to be addressed to make the measurement meaningful, accurate, and efficient. Another important feature is the population for which the measure is intended. This decision is not entirely dependent on the theoretical paradigm but more to the immediate research question at hand. 6/14/2016 2 A third point that needs mentioning is the purpose of the scale or measure. What is it that the researcher wants to do with the measure? Is it developed for a specific study or is it developed with the anticipation of extensive use with similar populations? Once some of these decisions are made and a measure is developed, which is a careful and tedious process, the relevant questions to raise are \" how do we know that we are indeed measuring what we want to measure? \" since the construct that we are measuring is abstract, and \" can we be sure that if we repeated the measurement we will get the same result? \". The first question is related to validity and second to reliability. Validity and reliability are two important characteristics of behavioral measure and are referred to as …
939 citations
TL;DR: In this article, the reliability measurement of qualitative data is important to determine the reliability of the qualitative data used in the analysis of marketing and applied marketing research, and it is discussed in detail.
Abstract: Data based on qualitative judgments are prevalent in both academic research in marketing and applied marketing research. Reliability measurement of qualitative data is important to determine the st...
859 citations
Book•
28 Feb 1994
TL;DR: Performance, Reliability, And Unwanted Consequences, and the Need for Models of the Reliability of Cognition.
Abstract: Performance, Reliability, And Unwanted Consequences. The Need for Models of the Reliability of Cognition. The Nature of Human Reliability Assessment. Reliability. Scientific Issues. The Fundamentals of the Model. The Dependent Differentiation Method. Chapter Summaries. Discussion. References. Index.
718 citations
Book•
01 Jan 1994
TL;DR: The role of research in nursing research, roles, approaches and issues as discussed by the authors, the scientific approach to the research process, the evolution of research from science to practice, and the role of qualitative approaches to research.
Abstract: Part 1 Nursing research - roles, approaches and issues: the role of research in nursing the scientific approach to the research process the evolution of research - from science to practice. Part 2 The research process: overview of the research process the literature review the theoretical framework the problem statement and hypothesis introduction to design experimental and quasiexperimental designs nonexperimental designs qualitative approaches to research sampling legal and ethical issues data collection methods reliability and validity descriptive data analysis inferential data analysis analysis of the findings. Part 3 Critique and application: evaluating quantitative research evaluating qualitative research.
706 citations
TL;DR: It was concluded that the usefulness of both measurement techniques in instructional research needs to be investigated further.
Abstract: The results of two of our recent empirical studies were considered to assess the usefulness of subjective ratings and cardiovascular measures of mental effort in instructional research. Based on its reliability and sensitivity, the subjective rating-scale technique met the requirements to be useful in instructional research whereas the cardiovascular technique did not. It was concluded that the usefulness of both measurement techniques in instructional research needs to be investigated further.
630 citations
TL;DR: It is shown that care must be taken in choosing a suitable ICC with respect to the underlying sampling theory, and a decision tree is developed that may be used to choose a coefficient which is appropriate for a specific study setting.
Abstract: In general, intraclass correlation coefficients (ICC's) are designed to assess consistency or conformity between two or more quantitative measurements. They are claimed to handle a wide range of problems, including questions of reliability, reproducibility and validity. It is shown that care must be taken in choosing a suitable ICC with respect to the underlying sampling theory. For this purpose a decision tree is developed. It may be used to choose a coefficient which is appropriate for a specific study setting. We demonstrate that different ICC's may result in quite different values for the same data set, even under the same sampling theory. Other general limitations of ICC's are also addressed. Potential alternatives are presented and discussed, and some recommendations are given for the use of an appropriate method.
627 citations
TL;DR: Extensions entail formulating alternative measures of CAPITA to clarify the theoretical foundations of the construct, validating the latent-structure model on another data set, use of multiple informants for data collection, and exploring complex factor structures for the construct.
Abstract: In order to measures the extent to which information technology provides competitive advantages, the construct "Competitive Advantage Provided by an Information Application" CAPITA was operationalized. A field survey gathered data from 185 top information systems executives regarding information technology applications which had been developed to gain competitive advantage. A confirmatory analysis revealed that CAPITA may be conceptualized in terms of nine dimensions which satisfy key measurement criteria including unidimensionality and convergent validity, discriminant validity, predictive validity, and reliability.
The nine dimensions from the basis of a preliminary multidimensional measure or index of competitive advantage which has practical uses for competitive assessment. These include justifying and evaluating applications and acting as dependent variables in empirical competitive advantage research.
Extensions entail formulating alternative measures of CAPITA to clarify the theoretical foundations of the construct, validating the latent-structure model on another data set, use of multiple informants for data collection, and exploring complex factor structures for the construct.
TL;DR: It is recommended that when both interrater and intrarater reliability are being assessed, a repeated-measures design be used to take advantage of the increased precision gained by using all observations in the statistical analysis, and appropriate statistical tests, confidence intervals, and SEMs always be used in conjunction with the estimated reliability coefficients.
Abstract: Statistical methodology for the concurrent assessment of interrater and intrarater reliability is presented. Application of the methodology is illustrated with an example of one therapist using two goniometers repeatedly to measure knee joint angles. Methods for estimating the coefficients, testing hypotheses, constructing confidence intervals, and computing sample size requirements are provided. In addition, the calculation and clinical interpretation of the standard error of measurement (SEM) are discussed. It is recommended that (1) when both interrater and intrarater reliability are being assessed, a repeated-measures design be used to take advantage of the increased precision gained by using all observations in the statistical analysis, and (2) appropriate statistical tests, confidence intervals, and SEMs always be used in conjunction with the estimated reliability coefficients.
TL;DR: In this paper, the first-order, second-moment approach is explored and applied to the design of embankment dams to evaluate the relative contributions of uncertainties about different soil parameters to the reliability of the embankments.
Abstract: Formally probabilistic methods for the analysis of slope stability have had relatively little impact on practice. Many engineers are not familiar with probabilistic concepts, and it has been difficult to incorporate concepts of reliability into practice. Also, there is confusion over what reliability and probability of failure mean. The most effective applications of probabilistic methods are those involving relative probabilities of failure or illuminating the effects of uncertainties in the parameters. Attempts to determine the absolute probability of failure are much less successful. The paper describes how probabilistic descriptions of soil parameters can be derived from field and laboratory data and applied in stability analysis. The first-order, second-moment approach is explored and applied to the design of embankment dams. The example illustrates the relative contributions of uncertainties about different parameters to the reliability of the embankment. Reliability analysis is especially useful in establishing design values of factors of safety representing consistent risks for different types of failure.
TL;DR: This article explored a dialectic between psychometric and hermeneutic approaches to drawing and warranting interpretations of human products or performances, and expanded the range of viable high-stakes assessment practices to include those that honor the purposes that students bring to their work and the contextualized judgments of teachers.
Abstract: Reliability has traditionally been taken for granted as a necessary but insufficient condition for validity in assessment use. My purpose in this article is to illuminate and challenge this presumption by exploring a dialectic between psychometric and hermeneutic approaches to drawing and warranting interpretations of human products or performances. Reliability, as it is typically defined and operationalized in the measurement literature (e.g., American Educational Research Association [AERA], American Psychological Association, & National Council on Measurement in Education, 1985; Feldt & Brennan, 1989), privileges standardized forms of assessment. By considering hermeneutic alternatives for serving the important epistemological and ethical purposes that reliability serves, we expand the range of viable high-stakes assessment practices to include those that honor the purposes that students bring to their work and the contextualized judgments of teachers.
TL;DR: In this article, the reliability estimation is limited to first-order reliability methods (FORM) for both component and systems reliability evaluation, and the solution strategies applying first order nonlinear optimization algorithms are described in detail with special attention to sensitivity analysis and stability of the optimization process.
TL;DR: The results show that the test-retest reliability is high and the validity of theBBT is shown by significant correlations between the BBT, an upper limb performance measurement and a functional independence measurement, which will help rehabilitation clinicians to differentiate better between real difficulties and those that may be attributed to normal aging.
Book•
01 Apr 1994
TL;DR: In this article, the main probabilistic methods employed in reliability and risk assessment, particularly fault tree analysis and failure mode, and effective analysis and their derivatives, are described and discussed.
Abstract: Fully updated and revised with more practical emphasis and more extensive use of case studies, this edition provides a comprehensive description of the main probabilistic methods employed in reliability and risk assessment, particularly fault tree analysis and failure mode, and effective analysis and their derivatives. Self-contained, each chapter focuses on a single subject or method and illustrates it with worked examples. The final chapter features four major case studies from recent projects to show how the methods have been applied to identify and quantify the risk and reliability characteristics against specific performance requirements. Focusing on the risk and reliability assessment of process plants, this book serves as a comprehensive reference to powerful analysis techniques for design, operation, and regulatory engineers involved directly with the safety and availability assessment of potentially hazardous process plants. It also serves as a text for undergraduate, graduate, and incompany courses and training. Contents: Introduction to Reliability and Risk Assessment; Reliability Mathematics; Qualitative Methods; Failure Mode and Effects Analysis; Quantific of Component Failure Probabilities; Reliability Networks; Fault Tree Analysis; Common Cause Failures; Maintainability; Markov Analysis; Simulation; Reliability Data Collection and Analysis; Risk Assessment; Case Study 1: Quantitative Safety Assessment of Ventilation Recirculation System in an Undersea Mine; Case Study 2: Failure Mode and Effects Criticality Analysis of Gas Turbine System; Case Study 3: In-Service Inspection of Structural Components (application to conditional maintenance of steam generators); Case Study 4: Business-interruption Risk Analysis. Appendix A: Expectation, Variance, and Standard Deviation. Appendix B: Standard Laplace Transforms; Glossary.
TL;DR: In this paper, the authors proposed an adaptive importance sampling (AIS) method that can be used to compute reliability and reliability sensitivities in turbine blade reliability analysis problems, and demonstrated the proposed AIS methodology is demonstrated using a turbine blade failure analysis problem.
Abstract: This paper presents recent developments in efficient structural reliability analysis methods. The paper proposes an efficient, adaptive importance sampling (AIS) method that can be used to compute reliability and reliability sensitivities. The AIS approach uses a sampling density that is proportional to the joint PDF of the random variables. Starting from an initial approximate failure domain, sampling proceeds adaptively and incrementally with the goal of reaching a sampling domain that is slightly greater than the failure domain to minimize over-sampling in the safe region. Several reliability sensitivity coefficients are proposed that can be computed directly and easily from the above AIS-based failure points. These probability sensitivities can be used for identifying key random variables and for adjusting design to achieve reliability-based objectives. The proposed AIS methodology is demonstrated using a turbine blade reliability analysis problem.
27 Jun 1994
TL;DR: Reliability theory applies to random variations in a population of similar objects, whereas software defects are all design flaws, not at all random, in a unique object.
Abstract: Reliability refers to statistical measures an engineer uses to quantify imperfection in practice. Often we speak imprecisely of an object having "high reliability", but technically, unless the object cannot fail at all, its reliability is arbitrarily close to zero for a long enough period of operation. This is merely an expression of the truism that an imperfect object must eventually fail. At first sight, it seems that software should have a sensible reliability, as other engineered objects do. But the application of the usual mathematics is not justified. Reliability theory applies to random (as opposed to systematic) variations in a population of similar objects, whereas software defects are all design flaws, not at all random, in a unique object. The traditional cause of failure is a random process of wear and tear, while software is forever as good (or as bad!) as new. However, software defects can be thought of as lurking in wait for the user requests that excite them, like a minefield through which the user must walk. >
TL;DR: Non-probabilistic convex models of uncertainty are used to formulate reliability in terms of acceptable system performance given uncertain operating environment or uncertain geometrical imperfections.
TL;DR: An experiment was carried out using trained observer-raters to evaluate the levels of drowsiness of drivers, the drivers' faces were recorded on videotape, and results indicate that observer rating are reliable and consistent.
TL;DR: The use of stochastic techniques is justified, and the various probability models that have been proposed, along with any associated statistical estimation and inference procedures, are described.
Abstract: Summary Probability models and statistical methods are a popular technique for evaluating the reliability of computer software. This paper reviews the literature concerning these methods, with an emphasis on the historical perspective. The use of stochastic techniques is justified, and the various probability models that have been proposed, along with any associated statistical estimation and inference procedures, are described. Examples of the models applied to real software failure data are given. A classic software development problem-how long software should be tested before it is released into the marketplace-is analyzed from a decision theoretic standpoint. Finally, the direction of future research is contemplated.
Proceedings Article•
01 Jan 1994
TL;DR: The authors describe the development, reliability testing, and validation of a 17-item instrument that measures the level of functioning of chronically mentally ill persons living in the community and predicts subsequent state and local hospital admissions.
Abstract: The authors describe the development, reliability testing, and validation of a 17-item instrument that measures the level of functioning of chronically mentally ill persons living in the community The Multnomah Community Ability Scale is designed to be completed by case managers The instrument provides a measure of the consumer's severity of disability which can, in turn, be used to: (a) describe an agency's "case mix" of clients; (b) measure consumer progress; (c) assign clients to different levels of service; and (d) assist payors in determining reimbursement The Multnomah Community Ability Scale is aimed specifically at persons with chronic mental illness, is sensitive to differences among individuals within this special population, and is quick and easy to complete The scale's reliability and validity have been examined in detail Inter-rater and test-retest reliability are good Criterion variables such as length of psychiatric hospitalization and clinicians' global ratings correlated highly with scale scores Finally, the instrument predicts subsequent state and local hospital admissions
Journal Article•
TL;DR: It is concluded that observation methods will have a place in epidemiologic studies on musculoskeletal disorders because of improvements and there have been improvements in the methods devised during the past 15 years.
Abstract: Establishing an association between physical work load and work-related musculoskeletal disorders requires a quantitative assessment of exposure to the main risk factors. Commonly used methods are questionnaires, diaries, interviews, systematic observations, and direct measurements. While questionnaires provide subjective information and have low reliability, measurements are technically sophisticated, expensive, and unable to identify some important risk factors. Systematic observations offer a compromise. The observation methods devised during the past 15 years are described, together with their reliability and validity. Most of them use duration or frequency of certain postures or events as the main measure of exposure. Their ability to quantify the main risk factors posture, manual handling, and repetitive work differs. All available methods have drawbacks and have been used to a limited extent in epidemiologic studies. There have been improvements however, and it is concluded that observation methods will have a place in epidemiologic studies on musculoskeletal disorders.
Proceedings Article•
01 Jan 1994TL;DR: In this paper, the authors used structural equation modeling to validate the measuring instrument used to derive the Computerworld rankings of information system effectiveness and found that both content validity and construct validity are questionable.
Abstract: Information system effectiveness is an important phenomenon for both researchers and practitioners. Despite widespread interest, and the importance of the uses, there have been no efforts to validate Computerworld's Premier 100 rankings of information system effectiveness. This paper uses structural equation modeling in an attempt to validate the measuring instrument used to derive the Computerworld rankings.Alternative models for the measuring instrument are proposed. Using a reflective model, the findings raise doubts as to the reliability of the rankings, and both content validity and construct validity are also suspect. The reliability and validity are problematic because multiple indicators of the same construct must be homogeneous for it to make sense to combine them into a composite index. A solution to this problem is to represent information system effectiveness as a multidimensional construct, that is part of a causal model. Based on previous research in the area, suggestions are offered to improve the measuring instrument.
TL;DR: In this article, an extension of the well known Hasofer-Lind-Rackwitz-Fiessler algorithm is proposed for the inverse reliability problem, where the search space of the random variables is extended here to include the unknown parameter and a search direction and a merit function are introduced to guide the sequence of iteration points to the design point and the parameter solution.
Abstract: Consider a reliability problem with a limit-state function involving one unknown parameter. We seek to determine the unknown parameter such that a prescribed first-order reliability index is attained. In this paper, an efficient algorithm is developed for the solution of this inverse reliability problem. The algorithm is an extension of the well known Hasofer-Lind–Rackwitz-Fiessler algorithm used in reliability analysis, where the search space of the random variables is extended here to include the unknown parameter. A search direction and a merit function are introduced to guide the sequence of iteration points to the design point and the parameter solution in a balanced manner. The proposed method can be used in reliability-based design optimization, or in determining the admissible response threshold for a specified first-order probability, without performing repeated reliability analyses. Two examples are presented: one demonstrating the characteristics of the convergence sequence, and another describing application of the algorithm to finite-element reliability analysis of an elastoplastic plate.
Journal Article•
TL;DR: The reliability of the hospital registry is better than their reputation, according to a study published in the British Journal of Clinical Practice.
Abstract: [Reliability of the hospital registry : The diagnostic data are better than their reputation]