Reduce Data Quality Research Articles

Quick assessment of hopelessness: a cross-sectional study

BackgroundLengthy questionnaires reduce data quality and impose a burden on respondents. Previous researchers proposed that a single item ("My future seems dark to me") and a 4-item component of the Beck's Hopelessness Scale (BHS) can summarise most of the information the BHS provides. There is no clear indication of what BHS cutoff values are useful in identifying people with suicide tendency.MethodsIn a population-based study of Chinese people aged between 15 and 59 in Hong Kong, the Chinese version of the BHS and the Centre for Epidemiologic Studies – Depression scale were administered by trained interviewers and suicidal ideation and suicidal attempts were self-reported. Receiver operating characteristics curve analysis and regression analysis were used to compare the performance of the BHS and its components in identifying people with suicidality and depression. Smoothed level of suicidal tendency was assessed in relation to scores on the BHS and its component to identify thresholds.ResultsIt is found that the 4-item component and, to a lesser extent, the single item of the BHS perform in ways similar to the BHS. There are non-linear relationship between suicidality and scores on the BHS and the 4-item component; cutoff values identified accordingly have sensitivity and specificity of about 65%.ConclusionThe 4-item component is a useful alternative to the BHS. Shortening of psycho-social measurement scales should be considered in order to reduce burden on patients or respondents and to improve response rate.

Feature enhancement and ATR performance using nonquadratic optimization-based SAR imaging

We present an evaluation of the impact of a recently proposed synthetic aperture radar (SAR) imaging technique on feature enhancement and automatic target recognition (ATR) performance. This image formation technique is based on nonquadratic optimization, and the images it produces appear to exhibit enhanced features. We quantify such feature enhancement through a number of criteria. The findings of our analysis indicate that the new feature-enhanced SAR image formation method provides images with higher resolution of scatterers, and better separability of different regions as compared with conventional SAR images. We also provide an ATR-based evaluation. We run recognition experiments using conventional and feature-enhanced SAR images of military targets, with three different classifiers. The first classifier is template based. The second classifier makes a decision through a likelihood test, based on Gaussian models for reflectivities. The third classifier is based on extracted locations of the dominant target scatterers. The experimental results demonstrate that the new feature-enhanced SAR imaging method can improve the recognition performance, especially in scenarios involving reduced data quality or quantity.

Cognitive procedures for correcting proxy‐response biases in surveys

AbstractTo reduce survey costs, major surveys rely on self‐ and proxy‐responses. The use of proxies can reduce data quality introducing biases in the survey estimates. This paper identifies one source of systematic differences between self‐ and proxy‐reports: proxies' higher reliance on inferences. Using data from the National Health Interview Survey on Disability (NHIS‐D), proxy‐response biases were modelled by independently collected measures of cognitive inferences. Conditional likelihood judgements about a number of disabilities (e.g. likelihood that a person has a disability given another disability) predicted the conditional disability reports for proxy‐ but not for self‐respondents (e.g. the proportion of respondents who reported difficulty learning after reporting difficulty communicating). A model of self/proxy differences was estimated on data from the 1994 NHIS‐D and tested against 1995 data. The correlation between predicted and actual differences was 0.76. The correlation between predicted and actual proxy‐reports was 0.95. Such research can be used to estimate and correct for systematic proxy‐response biases. Copyright © 2002 John Wiley & Sons, Ltd.

Perceptions of confidentiality protection at statistical agencies: Some evidence from data on businesses and households

This paper provides an overview of research into public perceptions of confidentiality. Statistical agencies expend a great deal of time and resources to protect the data that they collect from unauthorized disclosure and identification of individual responses. However, data protection, by its very nature, involves either reducing data quality or limiting access to the very information that statistical agencies go to so much trouble and expense to collect. Many statistical agencies know little about an important input into the data protection decision: the degree to which their own respondents - both businesses and households - understand and believe that statistical agencies have, in fact, delivered on their confidentiality promises and how such perceptions affect their responses. This paper provides a brief survey of the selected knowledge in the area. Results for the United States suggest that much work needs to be done to further this knowledge, and the paper argues that research into perceptions should be institutionalised by statistical agencies and used to inform data protection decisions. 1. Background Statistical agencies enter into an implicit pact with their respondents to protect the data with which they are entrusted - both because they are so mandated by law and because the protect is needed to maintain the trust necessary for high quality responses. In fulfilling this pact, they expend a great deal of effort applying different access protocols and data protection techniques. This protection comes at quite some cost - data protection, by its very nature, involves either reducing data quality or limiting access to the very information that statistical agencies go to so much trouble and expense to collect. However, it is ironic that many statistical agencies, whose role it is to provide information to make decisions, know little about an important input into the data protection decision: the degree to which their own respondents - both business and households - understand and believe that statistical agencies have, in fact, delivered on their confidentiality promises and how such perceptions affect their responses. This paper provides a survey of what is known about the public perceptions of how statistical agencies protect their data. 1 Note: This paper reports the results of research and analysis, part of which was undertaken by Census Bureau staff. It has undergone a Census Bureau review more limited in scope than that given to official Census Bureau publications. This report is released to inform interested parties of ongoing research and to encourage discussion of work in progress. All views expressed represent those of the authors, and not necessarily the institutions they represent.

Model comparisons for high-Z non-LTE steady-state calculations

Advances in computer technology and code development have made it possible to perform plasma kinetics calculations based on high- Z atomic structure corresponding to different levels of detail. Various atomic models have been implemented in the Los Alamos suite of codes in recent years for performing detailed non-equilibrium kinetics calculations. These include non-relativistic configuration average models with unresolved transition arrays (UTA), detailed fine structure models including configuration interaction and intermediate coupling that are capable of achieving spectroscopic accuracy for low- Z plasmas, and fully relativistic configuration average models employing the concept of the unresolved transition array, UTA, for modeling high- Z plasmas. Fully relativistic fine structure calculations have also been implemented and used primarily for accuracy checks. In addition, the fractional occupation number method has been introduced in the relativistic structure code to reduce computational times without reducing data quality. Finally, we will discuss a method that has been developed to use the populations extracted from a non-relativistic kinetics calculation in a relativistic spectral simulation. The purpose of the present work is to compare the emission spectra predicted by the various models and methods using consistent sets of atomic electron configurations and cross-sections for a gold plasma.

Violating Conversational Conventions Disrupts Cognitive Processing of Attitude Questions

This research distinguishes conversational norms from conversational conventions and tests the notion that violation of conversational conventions in attitude questions disrupts processing and reduces data quality. Our first study showed that in questions with simple, dichotomous affirmative and negative response alternatives, respondents expect the affirmative response alternative to be offered before the negative one. Four studies showed that violating this convention disrupts cognitive processing. Respondents took longer to answer questions asked unconventionally, and their answers were less predictable by their attitudinal dispositions and their demographic characteristics. In addition, answers were less responsive to manipulation of a key feature of an object of judgment. Another study ruled out a possible alternative explanation for the effect (an increase in disconfirmatory thinking) and documented that convention violation led people to generate more thoughts irrelevant to the question's topic. These disruption effects were reliable only among people for whom the cognitive work entailed by generating optimal answers to questions was most difficult.

Clinical experiences with the "Swedish interactive threshold algorithm" (SITA)

The "Swedish Interactive Threshold Algorithm" (SITA) is a new technique minimizing test time without reduction of data quality. In this retrospective study we compared visual fields assessed with the conventional full-threshold technique with those by SITA and analyzed qualitative and quantitative differences and the test time reduction of SITA. 113 eyes with different glaucomatous field damage of 66 patients aged from 22 to 89 years were included. The patients underwent perimetry with each strategy within four months. Exclusion criteria were eyes with progressing glaucomatous damage and high rate of false-positive results or artifacts by corrective glasses. The qualitative differences of the analysis were small: full threshold demonstrated more relative scotomas in eyes with normal SITA readings, whereas scotomas assessed with SITA were often pronounced to absolute scotomas. Average time reduction by SITA was 40% and depended on the severity of glaucomatous stage. No reduction was found for advanced glaucoma, whereas normal fields using SITA were performed in half of the time of full threshold strategy. Differences of SITA to conventional full threshold testing may be related to fatigue effects of full threshold strategy due to longer test time. The reduction of test time enables more motivation for more frequent visual field examinations and thus a better detection of early glaucoma or progressing visual field damage.

A new generation of algorithms for computerized threshold perimetry, SITA.

The purpose of this work was to develop a new family of test algorithms for computerized static threshold perimetry which significantly reduces test time without any reduction of data quality. A comprehensive visual field model constructed from available knowledge of normal and glaucomatous visual fields is continuously updated during testing. The model produces threshold estimates and also estimates of the certainty to which the threshold is known at each point. Testing is interrupted at each test location at predetermined levels of threshold certainty. New time-saving methods are employed for estimation of false answers, and test pacing is optimized. After completion of the test, all threshold estimates are re-computed, taking into account the complete body of patient responses. Computer simulations were used to optimize the different parameters of the new algorithms, to evaluate the relative importance of those parameters, and to evaluate the performance of the algorithm as a whole in comparison with a standard algorithm. Simulated test results obtained with this algorithm were slightly more accurate than those of the Humphrey Full Threshold test algorithm. The number of simulated stimuli presented was reduced by an average of 29% in normal fields and 26% in glaucomatous fields. Actual clinical test time should be further reduced, since the influence of the improved timing algorithm was not included in the simulations. We applied new methods which take available knowledge of visual field physiology and pathophysiology into account, and employ modern computer-intensive mathematical methods for real time estimates of threshold values and threshold error estimates. In this way it was possible to design a family of testing algorithms which significantly reduced perimetric test time without any loss of quality in results.

Automation of matrix-assisted laser desorption/ionization mass spectrometry using fuzzy logic feedback control.

Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry is a sensitive and versatile method for biomolecular analysis which has potential for high-throughput screening in many applications. To obtain mass spectra of optimal quality, however, laser fluence is continuously adjusted during data acquisition to be close to the threshold level of ion production, requiring a skilled operator and several minutes of acquisition time per sample. Using real-time fuzzy logic control of the laser fluence, we here demonstrate that the acquisition of MALDI spectra can be automated without reduction of data quality. The control algorithm evaluates signal intensity and mass resolution of the base peak. It then regulates the laser fluence to keep the ion signal intensity within the dynamic range of the data acquisition hardware while maintaining high mass resolution. This fuzzy logic control system allows unattended data acquisition using either static ion extraction or delayed ion extraction MALDI. Even for difficult samples such as femtomole-level peptide mixtures, no significant reduction in data quality is observed, as compared to manually obtained spectra. Automated analysis of 78 chromatographic fractions with high mass accuracy demonstrates the utility of the method. The control algorithm has been combined with other software modules to completely automate database identification of proteins by their peptide mass maps. The success of fuzzy logic in MALDI automation suggests wider uses of this technique in mass spectrometry.

Acoustic correlation sonar for vertical profiling of ocean currents to a range of 1 km

Ocean current profilers, based on spatial or temporal correlation analysis of sound backscattered from passive targets drifting with the current, have several potential advantages over the more common Doppler profilers. Correlation profilers can operate over a longer range with smaller, lighter and cheaper transducers. However, the signal processing is more complex and computationally intensive, increasing transmission intervals and leading to difficulties understanding when and why the performance is poor. Using data gathered from RRS Discovery in early 1995 the authors describe the performance of a 22 kHz acoustic correlation current profiler and compare its results with a lowered acoustic Doppler profiler. At low ship speeds, the signal to noise ratio was sufficient to allow profiling to 1 km depth, with the achievable range reducing as speed increased. Bias and decorrelation occurred in the upper 160 m owing to unknown causes and greatly reduced data quality.