Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistics”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,711 records · Page 95Linked to original sources
PubMed2002

Statistical techniques for the analysis of change in longitudinal studies of the menopause.

OBJECTIVE: Critical review of statistical techniques used to analyze data from longitudinal studies. METHOD: Literature search and classification of statistical methods and their utilization evaluated against known underlying assumptions. RESULTS: One hundred and twenty-three papers found: 1. cross-sectional reduction of data (56%); 2. multifactorial techniques (26%); 3. repeated measurement analysis of variance (14%); 4. other (time series and structural equation modelling) (4%). CONCLUSIONS: Cross-sectional reduction violates underlying statistical assumptions. A simple and powerful technique is to mean values prior to and following an event such as the final menstrual period. To allow for the influence of multiple factors, linear regression is preferred to logistic regression where continuous data are available. For more information about evolution in time, more complex techniques are needed. A suitable technique is repeated measurement multivariate analysis of variance using a number of contrasts to estimate various effects. Split plot or randomized block designs cannot be recommended as they violate compound symmetry assumptions. For series involving more than 100 observations for each subject, time series and spectral analysis techniques should be considered. Structural equation modelling is recommended for examination in detail of a range of factors that may influence the studied end-point, the presence of feedback and of latent or non-measurable variables.

Adult↗
PubMed2006

Statistical power and parameter stability when subjects are few and tests are many: comment on Peterson, Smith, Martorana, and Owens (2003).

Comments on the original article "The impact of chief executive officer personality on top management team dynamics: One mechanism by which leadership affects organizational performance", by R. S. Peterson et al.. This comment illustrates how small sample sizes, when combined with many statistical tests, can generate unstable parameter estimates and invalid inferences. Although statistical power for 1 test in a small-sample context is too low, the experimentwise power is often high when many tests are conducted, thus leading to Type I errors that will not replicate when retested. This comment's results show how radically the specific conclusions and inferences in R. S. Peterson, D. B. Smith, P. V. Martorana, and P. D. Owens's (2003) study changed with the inclusion or exclusion of 1 data point. When a more appropriate experimentwise statistical test was applied, the instability in the inferences was eliminated, but all the inferences become nonsignificant, thus changing the positive conclusions.

Data Interpretation, Statistical↗
PubMed1992

Can test statistics in covariance structure analysis be trusted?

Covariance structure analysis uses chi 2 goodness-of-fit test statistics whose adequacy is not known. Scientific conclusions based on models may be distorted when researchers violate sample size, variate independence, and distributional assumptions. The behavior of 6 test statistics is evaluated with a Monte Carlo confirmatory factor analysis study. The tests performed dramatically differently under 7 distributional conditions at 6 sample sizes. Two normal-theory tests worked well under some conditions but completely broke down under other conditions. A test that permits homogeneous nonzero kurtoses performed variably. A test that permits heterogeneous marginal kurtoses performed better. A distribution-free test performed spectacularly badly in all conditions at all but the largest sample sizes. The Satorra-Bentler scaled test statistic performed best overall.

Female↗
PubMed1983

Mathematical and statistical aids to evaluate data from renal patients.

The monitoring of renal patients and the making of many decisions during their management involves consideration of sequences of numerical data. Renal function results after renal transplantation were used as an example of how graphical presentations, simple mathematical transforms, statistical evaluation and adjustments to the data, to take into account other biological and technical sources of error, can all contribute to better understanding. Experience with a statistical technique, the 4-state Kalman filter not often used in the biological sciences, was summarized and its use suggested as a method to quantitate some traditionally subjective decisions about individual patients, for example, the onset of allograft rejection. The method has identified in retrospect and in prospect events after transplantation earlier than did experienced clinicians. Other statistical techniques to set the sensitivity and specificity of monitoring methods, to detect change points and to quantitate rhythmic sequences of clinical data were discussed, with examples, and with increasing access to computers, these can be used more easily by nephrologists, transplant surgeons, and others.

Computers↗
PubMed2006

Statistical practice in high-throughput screening data analysis.

High-throughput screening is an early critical step in drug discovery. Its aim is to screen a large number of diverse chemical compounds to identify candidate 'hits' rapidly and accurately. Few statistical tools are currently available, however, to detect quality hits with a high degree of confidence. We examine statistical aspects of data preprocessing and hit identification for primary screens. We focus on concerns related to positional effects of wells within plates, choice of hit threshold and the importance of minimizing false-positive and false-negative rates. We argue that replicate measurements are needed to verify assumptions of current methods and to suggest data analysis strategies when assumptions are not met. The integration of replicates with robust statistical methods in primary screens will facilitate the discovery of reliable hits, ultimately improving the sensitivity and specificity of the screening process.

Biological Assay↗
PubMed2001

Statistical methods for the analysis and presentation of the results of bone marrow transplants. Part I: unadjusted analysis.

In this paper, we describe modern statistical methods for presentation of the results of studies of bone marrow transplantation. We focus here on 'univariate' or unadjusted techniques to describe the outcomes of such studies. In another paper we will discuss multivariate methods. We discuss the type of data one may have available to make inference about outcomes. We explain the differences between the Kaplan-Meier estimator of the survival function and the cumulative incidence curve, how these curves should be interpreted and when each is the appropriate summary statistic. We discuss the weighted log rank statistic and show how different weights can be used to put emphasis on detecting differences between groups in different time periods. We also present a simple estimate of current leukemia-free survival which is useful in summarizing post-transplant events.

Analysis of Variance↗
PubMed2006

Simple allelic-phenotype diversity and differentiation statistics for allopolyploids.

The analysis of genetic diversity within and between populations is a routine task in the study of diploid organisms. However, population genetic studies of polyploid organisms have been hampered by difficulties associated with scoring and interpreting molecular data. This occurs because the presence of multiple alleles at each locus often precludes the measurement of genotype or allele frequencies. In allopolyploids, the problem is compounded because genetically distinct isoloci frequently share alleles. As a result, analysis of genetic diversity patterns in allopolyploids has tended to rely on the interpretation of phenotype frequencies, which loses information available from allele composition. Here, we propose the use of a simple allelic-phenotype diversity statistic (H') that measures diversity as the average number of alleles by which pairs of individuals differ. This statistic can be extended to a population differentiation measure (F'ST), which is analogous to FST. We illustrate the behaviour of these statistics using coalescent computer simulations that show that F'ST behaves in a qualitatively similar way to FST, thus providing a useful way to quantify population differentiation in allopolyploid species.

Alleles↗
PubMed1994

Inferential statistics and treatment efficacy studies in communication disorders.

A reconsideration of the dependence on inferential statistical analyses in treatment efficacy studies in communication disorders is suggested. The argument is developed that inferential statistical analyses as part of group research designs do not provide the kinds of data that are useful in determining the efficacy of clinical intervention. Furthermore, misinterpretations of the principles that underlie inferential statistics contribute to the difficulties in building a research base that can be applied to clinical issues. The view is taken that replication studies provide a more fruitful approach to ascertaining the efficacy of clinical intervention.

Communication Disorders↗
PubMed1994

Achieving statistical power through research design sensitivity.

The challenge for nurse researchers is to design their intervention studies with sufficient sensitivity to detect the treatment effects they are investigating. In order to meet this challenge, researchers must understand the factors that influence statistical power. Underpowered studies can result in a majority of null results in a research area when, in fact, the interventions are effective. The sensitivity of a research design is not a function of just one element of the design but of the entire research design: its plan, implementation and statistical analysis. When discussing factors that can increase a research design's statistical power, attention is most often focused on increasing sample size. This paper addresses a variety of factors and techniques, other than increasing sample size, that nurse researchers can use to enhance the sensitivity of a research design so that it can attain adequate power.

Nursing Research↗
PubMed2002

Statistical techniques for comparing measurers and methods of measurement: a critical review.

1. Clinical and experimental pharmacologists and physiologists often wish to compare two methods of measurement, or two measurers. 2. Biostatisticians insist that what should be sought is not agreement between methods or measurers, but disagreement or bias. 3. If measurements have been made on a continuous scale, the main choice is between the Altman-Bland method of differences and least products regression analysis. It is argued that although the former is relatively simple to execute, it does not distinguish adequately between fixed and proportional bias. Least products regression analysis, although more difficult to execute, does achieve this goal. There is almost universal agreement among biostatisticians that the Pearson product-moment correlation coefficient (r) is valueless as a test for bias. 4. If measurements have been made on a categorical scale, unordered or ordered, the most popular method of analysis is to use the kappa statistic. If the categories are unordered, the unweighted kappa statistic (K) is appropriate. If the categories are ordered, as they are in most rating scales in clinical, psychological and epidemiological research, the weighted kappa statistic (K(w)) is preferable. But K(w) corresponds to the intraclass correlation coefficient, which, like r for continuous variables, is incapable of detecting bias. Simple techniques for detecting bias in the case of ordered categorical variables are described and commended to investigators.

Bias↗
PubMed2000

Statistical maps of cerebral blood flow deficits in Alzheimer's disease.

Images of cerebral blood flow or metabolism are useful as adjunct to the differential diagnosis of cortical dementia. The aim of this study was to create statistical objective voxel maps of significant differences in regional cerebral blood flow between patients with Alzheimer's disease and age-matched healthy volunteers. Maps of significantly reduced cerebral blood flow were created based on a spatially normalized distribution of cerebral blood flow, measured with O-15-water and positron emission tomography in 16 Alzheimer's patients, compared to 16 healthy age-matched volunteers. After spatial normalization of voxel counts, the t-statistic of the cerebral blood flow deficit was determined from the local voxel-SDs. In the patients, significant reduction (P < 0.05) of the flow distribution was present in regions near the hippocampus, extending rostrally to the temporo-parietal region in both hemispheres, including the medial parietal cortex plus smaller frontal areas. The maximum reduction occurred in the left tapetum/hippocampus (53%, P = 0.061). In conclusion, statistical maps of cerebral blood flow deficits objectively reveal the location of deficits, identifying areas that are difficult to identify by subjective visual inspection of conventional sections of cerebral blood flow maps. This is particularly well illustrated by the pronounced flow reduction of the medial parietal cortices.

Aged↗
PubMed2004

A note on the genome scan meta-analysis statistic.

Wise and colleagues (Ann. Hum. Genet. (1999) 63: 263-72) introduced a rank-based statistical technique for meta-analysis of genome scans, the Genome Scan Meta-Analysis (GSMA) method. We provide an alternative derivation of the null distribution of the GSMA statistic, with extensions, and we suggest approximations to the distribution of the GSMA statistic that may be useful in applications.

Chromosome Mapping↗
PubMed2002

Vital statistics as a data source.

A focus group convened at a National Institutes of Health (NIH) Stillbirth Conference discussed issues related to vital statistics as a data source on fetal mortality. Fetal death is a tragic occurence and part of the continuum of pregnancy outcomes. A primary source of data is the National Vital Statistics System fetal death component, which is routinely used to track mortality trends and to provide the context for nonrepresentative investigations. The data are also used to examine differentials in mortality by characteristics and to evaluate data quality. Issues concerning how to improve the vital statistics data include culture and context; data instrument; persons providing information; what information is available to the information provider; processing issues; and use of data. Perhaps most important is to enhance recognition of fetal death both for the individual confronting a loss and to promote research. Other studies are needed to augment the information obtainable through vital records.

Culture↗
PubMed2003

Logit50: a nonlinear transformation of pNN50 with improved statistical properties.

The traditional time domain heart rate variability index pNN50 is a percentage scale-based measure of large beat-to-beat changes in heart period that may reflect parasympathetic neural activity impinging on the sino-atrial node. However, pNN50 exhibits nonlinear saturation effects near 0% and 100% that may adversely affect its statistical properties. The purpose of this paper is to propose a revision of pNN50, Logit50, that is the natural logarithm of the odds of the occurrence of large beat-to-beat differences in R-R interval. Using five clinical and normal sample data sets, the revised Logit50 index is shown to retain the computational simplicity and interpretability of the pNN50, but to have better metric properties in statistical and clinical applications. In particular, the Logit50 is demonstrated to be relatively unaffected by the positive distributional skew that is common in most statistical applications of pNN50.

Heart Rate↗
PubMed1977

[Quantification and statistical analysis of ordinal scale type data in psychiatry (author's transl)].

Observational data from psychiatric patients are usually of the ordinal scale type. Therefore, in the traditional one-rating-per-symptom-and-patient-situation, the resulting score values can be analysed only by applying nonparametric statistical methods. These methods are rather inefficient when--as is usual--there are only four-point-scales for evaluating the symptoms. Moreover, non-parametric methods are not available as of yet to take into consideration the often complex structure of data from psychiatric studies. This leads to the conclusion that the quantification of ordinal scale type data would be very desirable in order that parametric statistical methods could be applied. Such a method of quantification is proposed on the basis of the audiovisual psychiatric rating technique. It is shown that the data resulting from the many-ratings-per-symptom-and-patient-situation can be subjected to parametric statistical methods. In applying the proposed method, therefore, much better use could be made of and a maximum of information could be extracted from the data in psychiatric studies.

Humans↗
PubMedDate not supplied

[What to do if statistical power is low? A practical strategy for pre-post-designs].

This article deals with the issue of statistical validity when evaluating interventions. The most common study design with two groups and two points of measurement is discussed. In clinical research settings, unsatisfactory statistical validity is often seen due to small sample sizes. In order to resolve this problem, a strategy based on an approach by Hager is proposed which takes both significance testing and effects sizes systematically into account. Using an example from clinical research practice the problematic issue of statistical power is introduced and methods to increase the power of tests are discussed. Within this framework, Erdfelder's compromise power analysis (computing alpha levels according to a predetermined beta/alpha error ratio) is crucial as well as a lowering of the number of applied tests by data reduction and the improved detection of potential effects by methods to reduce error variance. The results show that significance tests should not be used in case of small sample and effect sizes. In these cases different approaches should be used.

Data Interpretation, Statistical↗
PubMed2004

[Care required by disabled persons: are official severe disability statistics good enough for requirement analyses?].

The appropriateness of the German Official Severe Disability Statistics for determining the amount of care needed by disabled persons is a controversial issue and therefore in the focus of this article. The criteria for determining the care required by disabled persons are pointed out. Data on prevalence and incidence exemplify that Official Severe Disability Statistics is only partially appropriate for assessing future care requirements. The problems become particularly obvious in topics of data collection, data processing and data storage. The deletion of the statistical data at regular intervals of five years makes long term prognosis almost impossible. While the problems concerning data processing and data storage could be solved by few modifications, the improvement of data collection could only be achieved by including and applying the International Classification of Functioning (ICF) of the WHO.

Activities of Daily Living↗
PubMed1985

Statistics in practice. Comparing the means of several groups.

This article discusses statistical methods for comparing the means of several groups and focuses on examples from 50 Original Articles published in the Journal in 1978 and 1979. Although medical authors often present comparisons of the means of several groups, the most common method of analysis, multiple t-tests, is usually a poor choice. Which method of analysis is appropriate depends on what questions the investigators wish to ask. If the investigators want to identify which of the groups under study are different from the rest, they will need a different method from the one required if they wish simply to decide whether or not the groups share a common mean. More complicated questions about the group means call for more sophisticated techniques. Of the 50 Journal articles examined, 27 (54 per cent) used inappropriate statistical methods to analyze the differences between group means. Investigators need to become better acquainted with statistical techniques for making multiple comparisons between group means.

Analysis of Variance↗