Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Sample Size”

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 541 records · Page 30Linked to original sources
PubMed2002

Estimating sample size for tests on trends across repeated measurements with missing data based on the interaction term in a mixed model.

A formula to estimate the required sample size for a study with repeated measurements was constructed based on the test of an interaction term in a mixed model. It covers both random effects and serial correlation and allows for missing data. This formula indicates that the method suggested by Dawson is conservative. A simulation study further verified the accuracy of the formula. Factors that influence the required sample size, such as the number of repeated measurements, the structure of the within-subject correlation, and their interaction, were investigated in detail.

Bias↗
PubMed1991

Sample sizes and power computation for clinical intervention trials.

The purpose of this article was to describe basic concepts of sample size and power estimation for planning nursing intervention trials and interpreting their results. Simple mathematical calculations, using the formulas presented here, can be used to estimate the number of subjects required to conduct a study with a designated effect size and level of power. These methods are of great importance, since most funding agencies require sample size and power estimations before a grant is awarded. In general, studies with power lower than .7 or .8 need careful consideration before they are implemented. In these situations, it may be wise to consider various alternatives for obtaining study subjects or deleting treatment groups for investigations involving more than two groups. The formulas presented here can also be useful in estimating the power of published research findings. Through a quick calculation, the consumer of nursing research can critically evaluate the meaning of a negative trial and draw appropriate conclusions for future research and practice.

Clinical Nursing Research↗
PubMed2001

On sample size and inference for two-stage adaptive designs.

Proschan and Hunsberger (1995, Biometrics 51, 1315-1324) proposed a two-stage adaptive design that maintains the Type I error rate. For practical applications, a two-stage adaptive design is also required to achieve a desired statistical power while limiting the maximum overall sample size. In our proposal, a two-stage adaptive design is comprised of a main stage and an extension stage, where the main stage has sufficient power to reject the null under the anticipated effect size and the extension stage allows increasing the sample size in case the true effect size is smaller than anticipated. For statistical inference, methods for obtaining the overall adjusted p-value, point estimate and confidence intervals are developed. An exact two-stage test procedure is also outlined for robust inference.

Biometry↗
PubMed2003

Evaluation of a digestion assay and determination of sample size and tissue for the reliable detection of Trichinella larvae in walrus meat.

A digestion assay was validated for the detection of Trichinella larvae in walrus (Odobenus rosmarus) meat, and appropriate samples for testing were determined using tissues from infected walruses harvested for food. Examination of muscles from 3 walruses showed that the tongue consistently contained approximately 2-6 times more larvae than the pectoral and intercostal muscles. Comparison of numbers of larvae in the root, body, and apex of the tongue from 3 walruses failed to identify a predilection site within the tongue, but the apex was considered an optimal tissue because of the high larval density within the tongue and the ease of collection. All 31 spiked samples weighing 50 g each and containing between 0.1 and 0.4 larvae per gram (lpg) were correctly identified as infected, indicating that the sensitivity of this procedure is adequate for diagnostic use. A sample size of 10 g consistently detected larvae in 2 walrus tongues containing > or = 0.3 lpg (n = 40), and until additional data are available, sample sizes from individual walrus tongues should be a minimum of 10 g. This study provides the preliminary data that were used for the development of a food safety analytical protocol for the detection of Trichinella in walrus meat in arctic communities.

Animals↗
PubMed2004

Technical variability and required sample size of helminth egg isolation procedures: revisited.

Mes [Vet Parasitol (2003) 115:311-320] recently reported a quantitative study of repeated measurements of nematode egg counts in faecal samples from dairy cattle, in order to compare the faecal egg counts resulting from two different laboratory techniques, the widely used McMaster method and a newer salt-sugar flotation (SSF) method. He concluded that the SSF technique requires much smaller sample sizes, and is also potentially simpler to carry out, making it the method of choice. Here I re-analyse these data to show that if the comparison is done on the most appropriate measurement scale (lognormal), and the large difference in multiplication factors is taken into account, then there is little to choose between the McMaster and SSF techniques as far as the required sample size is concerned. In particular, the treatment of the data as normally rather than lognormally distributed leads to incorrect statistical tests, power analyses and confidence intervals.

Analysis of Variance↗
PubMed2001

Sample size estimations for MRI-monitored trials of MS comparing new vs standard treatments.

The authors estimated the sample sizes needed for exploratory trials of MS assessing the efficacy of new treatments in reducing the number of new enhancing lesions vs those of interferon-beta or glatiramer acetate. The sample sizes per arm ranged from 868 (effect: 20%) to 94 (effect: 50%) for patients with relapsing-remitting MS and from 2,484 (effect: 20%) to 361 (effect: 50%) for patients with secondary progressive MS. In MS, exploratory trials of new vs available therapies require large numbers of patients, even when MR end-points are used.

Adjuvants, Immunologic↗
PubMed1999

Sample size determination for proving equivalence based on the ratio of two means for normally distributed data.

Equivalence trials aim to demonstrate that two treatments do not differ by more than a prespecified clinically irrelevant amount. We consider the problem when equivalence is defined in terms of the ratio of population means and the original (untransformed) data are normally distributed. Application of the intersection-union principle to the test proposed by Sasabuchi results in a two one-sided tests procedure of size alpha. We give the associated 100 (1-2 alpha) per cent confidence interval and derive the exact methods for calculation of power and sample sizes for the parallel group design and the two-period cross-over. We present tables and figures of required sample sizes and achieved power.

Administration, Inhalation↗
PubMed1993

Sample sizes for prevention trials have been too small.

Planners of several large prevention trials have overestimated the expected incidence of events in the control group, largely because they failed either to recognize or to adequately correct for various effects of population selection. Consequently, the studies have been too small in size or too short in duration to achieve their stated objectives. The selection effects include those engendered by the choice of the target population, the self-selection of volunteers, and protocol exclusions. This paper presents a taxonomy of these effects and the likely direction of their influence on the incidence of events and on mortality rates from other causes. Little information is available to help sample-size planners in adjusting for these effects. A few studies have provided information on the extent to which control group incidence rates have fallen short of expectations. In particular, researchers from the University of Minnesota's Colon Cancer Control Study have provided a detailed comparison of event incidence and all-cause mortality rates with general population rates. (AM J Epidemiol 1993;137:797-810). Other studies should publish similarly detailed information to assist sample-size planners of prevention trials. Until more information is published, this paper provides preliminary guidelines for prevention trial sample-size planning.

Clinical Trials as Topic↗
PubMedDate not supplied

Confidence intervals and sample-size calculations for the sisterhood method of estimating maternal mortality.

The sisterhood method is an indirect method of estimating maternal mortality that has, in comparison with conventional direct methods, the dual advantages of ease of use in the field and smaller sample-size requirements. This report describes how to calculate a standard error to quantify the sampling variability for this method. This standard error can be used to construct confidence intervals and statistical tests and to plan the size of a sample survey that employs the sisterhood method. Statistical assumptions are discussed, particularly in relation to the effective sample size and to effects of extrabinomial variation. In a worked example of data from urban Pakistan, a maternal mortality ratio of 153 (95 percent confidence interval between 96 and 212) deaths per 100,000 live births is estimated.

Age Factors↗
PubMed2000

On the relative sample size required for multiple comparisons.

Multiple comparisons are commonly made in epidemiologic and genetic research. How to appropriately adjust for multiple comparisons remains a controversial issue. This note demonstrates, however, that large increases in the number of comparisons has a limited effect on the sample size required to maintain an experimentwise alpha-level. In particular, the relative sample size required increases only linearly with the logarithm of the number of comparisons made.

Models, Statistical↗
PubMed1998

Determination of sample sizes for the estimation of Onchocerca volvulus (Filarioidea: Onchocercidae) infection rates in biting populations of Simulium ochraceum s.l. (Diptera: Simuliidae) and its application to ivermectin control programs.

Monthly samples of biting Simulium ochraceum s.l. Walker were collected before and after ivermectin treatment in southern Mexico and analyzed for Onchocerca volvulus Leuckart infection rates, infection intensity, and the characteristics of larval distribution among parous flies. The variance over mean ratio (VMR) indicated that in all cases this distribution departed from Poisson and was strongly aggregated (VMR > 1). The negative binomial was found to be an adequate model with a small value of the aggregation parameter k, but the degree of larval overdispersion increased as the mean larval load decreased, invalidating the use of a common kc value. A linear relationship between k and the mean (mu) was established, k(mu) = k1 mu, which permitted exploration of the relationship between the observed proportion of infected flies, p, and the estimated mean larval burden per fly, m (all larval stages in parous flies). This would allow mean numbers of larvae per parous fly to be predicted from presence-absence data (e.g., from infection rates provided by polymerase chain reaction methods applied to pools of flies), assuming that k1 is a known parameter. Given that both p and m are naturally low in S. ochraceum, their relationship was practically linear within the range of observed values. Predictions were tested with the Mexican data from which the clumping parameter was estimated as well as for Guatemalan data for which this information was not available. Results showed a highly satisfactory degree of agreement between predictions and observations. The sample sizes required to estimate mean larval loads from prevalence data for fixed levels of precision (defined as the ratio between SE[m] and m) were calculated for realistic S. ochraceum infection rates (those found in published pre- and postcontrol field surveys as well as in this work). For the special case in which the relationship between k and the mean is linear and goes through the origin, k(mu) = k1 mu, the number of flies to be examined for O. volvulus infections does not explicitly depend on the aggregation parameter, but rather on the unknown proportion of infected flies. Practical recommendations for the calculation of sample sizes are discussed. For infection levels < 0.2%, a minimum number between 6,000 and 13,000 parous flies would have to be examined to estimate the mean larval load with a precision between 0.20 and 0.30. The linearity between onchocercal infection rate and infection intensity in the fly population indicates that relationships between the former and onchocerciasis patterns in the human population should be further explored for the purposes of monitoring the impact of ivermectin control programs through entomological evaluations.

Animals↗
PubMed1985

[Random sample size in ROC analyses].

ROC analysis has proved useful in assessing diagnostic efficiency. Large x-ray film series of thoracic images intended for cardiovascular diagnosis were used for studying diagnostic efficiency in relation to random sample size. Exploration of random samples from two classes of findings requiring diagnostic differentiation, with the same number of x-ray films in each class, showed satisfactory convergence between the radiologists' rating and the ROC curve if the size of the random sample groups was about 200 images each or larger. The smaller the random sample size (less than 200 images), the greater the scatter of the rating performance. In another series the rating ability of experienced radiologists was compared with that of a semi-automatic classificator. The semi-automatic classificator attained almost the same performance index as the low-performance evaluator.

Decision Theory↗
PubMedDate not supplied

Critical sample sizes for determining the statistical significance of mutation frequencies.

Based on the assumption that the numbers of mutations observed in an untreated and treated sample of individuals are binomial random variables, a method is presented to compute the probability of observing a specific number of mutations as a function of the sample sizes and the number of mutations in the untreated control sample. Knowledge of the true mutation frequencies is not required. The formalism is then used to compute critical sample sizes for testing hypotheses concerning mutation frequencies in the two populations.

Mutagenicity Tests↗
PubMed2002

Exploratory factor analysis in behavior genetics research: factor recovery with small sample sizes.

Results of a Monte Carlo study of exploratory factor analysis demonstrate that in studies characterized by low sample sizes the population factor structure can be adequately recovered if communalities are high, model error is low, and few factors are retained. These are conditions likely to be encountered in behavior genetics research involving mean scores obtained from sets of inbred strains. Such studies are often characterized by a large number of measured variables relative to the number of strains used, highly reliable data, and high levels of communality. This combination of characteristics has special consequences for conducting factor analysis and interpreting results. Given that limitations on sample size are often unavoidable, it is recommended that researchers limit the number of expected factors as much as possible.

Animals↗
PubMed1999

Sample size requirements for precise estimates of reliability, generalizability, and validity coefficients.

Precision of the reliability coefficient (r) is investigated. The width of the confidence interval for r as a function of sample size (N) is shown for retest, alternate-form, split-half, alpha, intraclass, interrater, and validity coefficients. Although the determination of the N needed for reliability studies is somewhat subjective, a minimum of 400 subjects is recommended. Much larger Ns may be needed for validity studies. A survey of published reliability studies shows that 59% of the sample sizes were less than 100. Confidence intervals for obtained test scores are used as a practical application measure that also leads to the conclusion of a minimum of 400 subjects.

Bias↗
PubMed1990

Sample-size guidelines for linkage analysis of a dominant locus for a quantitative trait by the method of lod scores.

Sample-size guidelines for linkage studies of quantitative traits partially determined by a dominant major locus are needed to provide a rough estimate of the amount of pedigree material that should be sampled to map the loci that influence such traits. After pedigrees are sampled, a specific power calculation can be carried out to evaluate the linkage information provided by the sampled pedigrees. Using computer simulation, I provide sample-size guidelines for linkage studies by the method of lod scores of quantitative traits partially determined by a dominant major locus. I consider the effects of a trait model, marker characteristics, and sampling strategy, with particular attention to sampling strategy because it is the one factor which the investigator can fully control. My results suggest that linkage studies of quantitative traits are practical, particularly if the investigator chooses an efficient sampling design and an efficient strategy to select pedigrees for linkage analysis.

Female↗
PubMed2001

Design issues and sample size when exposure measurement is inaccurate.

Measurement error often leads to biased estimates and incorrect tests in epidemiological studies. These problems can be corrected by design modifications which allow for refined statistical models, or in some situations by adjusted sample sizes to compensate a power reduction. The design options are mainly an additional replication or internal validation study. Sample size calculations for these designs are more complex, since usually there is no unique design solution to obtain a prespecified power. Thus, additionally to a power requirement, an optimal design should also fulfill the criteria of minimizing overall costs. In this review corresponding strategies and formulae are described and appraised.

Bias↗