Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Sample size estimation”

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 685 records · Page 38Linked to original sources
PubMed2000

Statistics in ophthalmic research: two eyes, one eye or the mean?

BACKGROUND: Ophthalmic data, while different among individuals, are usually similar between fellow eyes of the same individual. This study was designed to illustrate alternative approaches to account for the correlation between fellow eyes. This is important for making inferences using data from both eyes. METHODS: With the use of a real data set from a population-based study, we described the distribution of intraocular pressure (IOP) by estimating the mean and standard deviation (SD) and evaluated the potential risk factors of higher IOP based on the regression method. The units of observation studied were of both eyes, right eye only, left eye only, the eyes with higher IOP and the mean value of both eyes. Furthermore, the generalized estimating equation (GEE) method was used to account for the correlation between fellow eyes in the regression analysis. Results and inferences from the different approaches were compared. RESULTS: The analysis included all the eyes, providing the largest sample size and unbiased estimates of the mean and SDs. There were some discrepancies among different approaches in the regression analysis. The GEE method simultaneously evaluated the effects of both eyes, and increased precision and enhanced inferences. CONCLUSIONS: Inconsistent results among different ophthalmic studies result from variations in not only study design and courses but also statistical methods. Making the best use of appropriate statistical techniques, which account for between eye correlation, provides valid statistical inferences.

Blood Pressure↗
PubMed2002

Sample size determination.

Scientists who use animals in research must justify the number of animals to be used, and committees that review proposals to use animals in research must review this justification to ensure the appropriateness of the number of animals to be used. This article discusses when the number of animals to be used can best be estimated from previous experience and when a simple power and sample size calculation should be performed. Even complicated experimental designs requiring sophisticated statistical models for analysis can usually be simplified to a single key or critical question so that simple formulae can be used to estimate the required sample size. Approaches to sample size estimation for various types of hypotheses are described, and equations are provided in the Appendix. Several web sites are cited for more information and for performing actual calculations

Animals↗
PubMed2007

A comparison of adaptive allocation rules for group-sequential binary response clinical trials.

In clinical trials to compare two or more treatments with dichotomous responses, group-sequential designs may reduce the total number of patients involved in the trial and response-adaptive designs may result in fewer patients being assigned to the inferior treatments. In this paper, we combine group-sequential and response-adaptive designs, extending recent work on sample size re-estimation in trials to compare two treatments with normally distributed responses, to analogous binary response trials. We consider the use of two parameters of interest in the group-sequential design, the log odds ratio and the simple difference between the probabilities of success. In terms of the adaptive sampling rules, we study two urn models, the drop-the-loser rule and the randomized Pólya urn rule, and compare their properties with those of two sequential maximum likelihood estimation rules, which minimize the expected number of treatment failures. We investigate two ways in which adaptive urn designs can be used in conjunction with group-sequential designs. The first method updates the urn at each interim analysis and the second method continually updates the urn after each patient response, assuming immediate patient responses. Our simulation results show that the group-sequential design, which uses the drop-the-loser rule, applied fully sequentially, is the most effective method for reducing the expected number of treatment failures and the average sample number, whilst still maintaining the nominal error rates, over a range of success probabilities.

Clinical Trials as Topic↗
PubMed1998

The significance of non-significance.

We discuss the implications of empirical results that are statistically non-significant. Figures illustrate the interrelations among effect size, sample sizes and their dispersion, and the power of the experiment. All calculations (detailed in Appendix) are based on actual noncentral t-distributions, with no simplifying mathematical or statistical assumptions, and the contribution of each tail is determined separately. We emphasize the importance of reporting, wherever possible, the a priori power of a study so that the reader can see what the chances were of rejecting a null hypothesis that was false. As a practical alternative, we propose that non-significant inference be qualified by an estimate of the sample size that would be required in a subsequent experiment in order to attain an acceptable level of power under the assumption that the observed effect size in the sample is the same as the true effect size in the population; appropriate plots are provided for a power of 0.8. We also point out that successive outcomes of independent experiments each of which may not be statistically significant on its own, can be easily combined to give an overall p value that often turns out to be significant. And finally, in the event that the p value is high and the power sufficient, a non-significant result may stand and be published as such.

Analysis of Variance↗
PubMed2003

Obtaining power or obtaining precision. Delineating methods of sample-size planning.

Sample-size planning historically has been approached from a power analytic perspective in order to have some reasonable probability of correctly rejecting the null hypothesis. Another approach that is not as well-known is one that emphasizes accuracy in parameter estimation (AIPE). From the AIPE perspective, sample size is chosen such that the expected width of a confidence interval will be sufficiently narrow. The rationales of both approaches are delineated and two procedures are given for estimating the sample size from the AIPE perspective for a two-group mean comparison. One method yields the required sample size, such that the expected width of the computed confidence interval will be the value specified. A modification allows for a defined degree of probabilistic assurance that the width of the computed confidence interval will be no larger than specified. The authors emphasize that the correct conceptualization of sample-size planning depends on the research questions and particular goals of the study.

Analysis of Variance↗
PubMed2001

Perchlorate levels in samples of sodium nitrate fertilizer derived from Chilean caliche.

Paleogeochemical deposits in northern Chile are a rich source of naturally occurring sodium nitrate (Chile saltpeter). These ores are mined to isolate NaNO3 (16-0-0) for use as fertilizer. Coincidentally, these very same deposits are a natural source of perchlorate anion (ClO4-). At sufficiently high concentrations, perchlorate interferes with iodide uptake in the thyroid gland and has been used medicinally for this purpose. In 1997, perchlorate contamination was discovered in a number of US water supplies, including Lake Mead and the Colorado River. Subsequently, the Environmental Protection Agency added this species to the Contaminant Candidate List for drinking water and will begin assessing occurrence via the Unregulated Contaminants Monitoring Rule in 2001. Effective risk assessment requires characterizing possible sources, including fertilizer. Samples were analyzed by ion chromatography and confirmed by complexation electrospray ionization mass spectrometry. Within a lot, distribution of perchlorate is nearly homogeneous, presumably due to the manufacturing process. Two different lots we analyzed differed by 15%, containing an average of either 1.5 or 1.8 mg g-1. Inadequate sample size can lead to incorrect estimations; 100-g samples gave sufficiently consistent and reproducible results. At present, information on natural attenuation, plant uptake, use/application, and dilution is too limited to evaluate the significance of these findings, and further research is needed in these areas.

Chile↗
PubMed1995

Speckle analysis using signal to noise ratios based on fractional order moments.

The SNR (signal-to-noise ratio) of the echo envelope image is a monotonically-increasing function of scatterer number density. Various SNRs, like amplitude SNR and intensity SNR, can be used to quantify the scatterer density. The problem of using a SNR based on higher order moments like the intensity SNR is that they require large sample sizes to obtain estimates with high confidence (the variance of the estimate becomes large for higher moments). In this paper, we consider SNRs based on fractional order moments (moments of order less than 1), and obtain mathematical analyses of their properties using the K distribution, which has been shown to be a good model for the density function of backscatter echo envelope signal. Statistics of SNRs based on fractional moment are derived and appear to be more robust and useful than the amplitude and intensity SNRs previously studied. The SNRs based on fractional order moments have greater dynamic range and the sample size requirements are smaller than those for integral order moment SNRs, like amplitude SNR or intensity SNR. Thus, SNRs based on fractional order moments could be used to better quantify the variations in scatterer density which can be used for tissue classification problems.

Humans↗
PubMed2004

Predicting sampling saturation of mtDNA haplotypes: an application to an enlarged Portuguese database.

An enlarged mtDNA database ( n=549) for the Portuguese population, comprising HVRI and HVRII regions is reported. This database was used to test the effect of sample size on the estimation of relevant parameters such as haplotype diversity, number of different haplotypes, nucleotide diversity and number of polymorphic positions. Simulations were performed generating sets of random subsamples of variable sizes ( n=50, 100, 200, 300 and 400). The results show that while haplotype and nucleotide diversities do not vary significantly with sample size, the numbers of haplotypes and polymorphic positions rise continuously inside the tested interval. These trends are interpretable by the evolution of the proportions of sequences that are found once or twice, which drop dramatically as sample size increases, with the corresponding rise in the frequency of those encountered 3 times or more. The generated data were also used to extrapolate saturation curves for the referred parameters. When considering for instance the number of haplotypes, it is shown that a sample size of 1,000 individuals is required for practical saturation (defined as the point where a sample size increase of 100 individuals corresponds to an increment in the diversity measure below 5%). For HVRII the same level is reached at n=900 and n=1,300 is needed when both regions are analysed simultaneously. Consequently, we can infer that currently used sample sizes are still rather inadequate for both anthropological and forensic purposes.

DNA, Mitochondrial↗
PubMed2003

Cost-effectiveness of a primary care intervention for depressed females.

BACKGROUND: To date, there is little information about the differential impact of primary care interventions by gender. We conducted an exploratory cost-effectiveness analysis by gender of an intervention to improve recognition and guideline-concordant treatment of depression in primary care. METHODS: Primary care practices that did not employ an onsite mental healthcare specialist were randomized to enhanced (intervention) versus usual care. All subjects met study criteria for current major depression. Medical Outcomes Study SF-36 scores were converted into quality-adjusted life years (QALYs) to compare the 1-year effectiveness of enhanced versus usual care by gender. Based on results of previous studies, antidepressant acceptors beginning a new depression treatment episode were the focus of the analysis. Statistical analyses included multivariate regression models controlling for sociodemographic and clinical covariates. RESULTS: In the main analysis, enhanced care for females was more expensive and more effective than usual care, at an additional cost of $5244 per QALY. For males, enhanced care was essentially cost and outcome neutral compared to usual care. The cost-effectiveness ratio estimates were robust to sensitivity analyses. Psychological side effects to the intervention may partially explain the limited effect of the intervention on outcomes for males. LIMITATIONS: We consider these results exploratory because the SF-36 to quality-adjusted life year conversion formula is preliminary and because of the relatively small sample size. CONCLUSIONS: The estimated cost-effectiveness ratio of this depression intervention is within the acceptable range for females, but not males. If replicated, these exploratory findings suggest that interventions to improve primary care depression treatment may need to be modified to improve their effectiveness in males while maintaining their effectiveness in females.

Adult↗
PubMed2006

Comparing spectra and coherences for groups of unequal size.

Spectra and coherences are standard measures of association within and between time series. These measures have several advantages over their time-domain counterparts, not the least of which is the ability to derive and estimate confidence intervals. However, comparing spectra and coherences between two groups of observation is a problem that has not received much attention. This problem is important in neuroscience since it is often of great interest to determine whether the estimates differ between distinct experimental/behavioral conditions. Here we propose one approach to this problem. Based on the known distributional properties of spectral and coherence estimates, we derive a test for equality of two spectral or coherence estimates. The test is applicable to unequal sample sizes. We also derive jackknifed estimates of the variance of the proposed test statistic. We suggest that comparing the estimates obtained from the jackknife procedure with the theoretical estimates provides a robust means of determining whether the data in question shows non-Gaussian or non-stationary behavior. Finally, we present applications of the method to simulated and real data.

Computer Simulation↗
PubMed1990

Intraspecific variation in the reproductive capacity of Anopheles stephensi (Diptera: Culicidae).

Intraspecific variation in the reproductive capacity of Anopheles stephensi Liston females was studied under constant laboratory conditions. Of 421 engorged females examined individually throughout their lifetimes, 260 laid a total of 479 egg batches with a maximum of nine ovipositions per female. The number of eggs per oviposition varied from 5 to 247 (average 96.8). The number of ovipositions per female were correlated positively with the average number of eggs per batch and exhibited a negative binomial distribution among females, indicating that a small portion of the population exhibited a particularly high fecundity. Among the ovipositing females, the total fecundity and fertility ranged, respectively, from 5 to 1,084 eggs and from 0 to 1,036 larvae per female. The estimated maximal number of female progeny produced by a female in her lifetime was 317, and the estimates of net reproductive rate (Ro) and capacity for increase (rc) were 25.7 and 0.17, respectively. The number of eggs in the first oviposition was predictive of the total fecundity of a female. The wide range of individual variation in the reproductive capacity of An. stephensi emphasized the requirement for a large sample size for reliable estimation.

Animals↗
PubMed1993

Extended multipoint identity-by-descent analysis of human quantitative traits: efficiency, power, and modeling considerations.

Goldgar introduced a novel marker-based method for partitioning the variation of a quantitative trait into specific chromosomal regions. Unlike traditional linkage mapping methods, Goldgar's method does not require the estimation of statistical quantities characterizing each locus thought to influence the trait under scrutiny (e.g., allele frequencies, penetrances, etc.). Goldgar's method is thus more flexible and less model dependent than many traditional marker-based genetic analysis techniques. Unfortunately, however, many of the properties of Goldgar's method have not been investigated. In this paper, the utility of an extended version of Goldgar's approach is studied in settings in which sibships are taken as the sampling unit of interest. The extensions discussed resolve around the incorporation of a wider variety of effects and factors into Goldgar's basic model. Analytic studies pertaining to power, sample-size requirements, and estimation procedures for the proposed extended version of Goldgar's method are described. Hypothesis-testing strategies are also discussed. The results of the analytic studies indicate that, although an extended sib-pair version of Goldgar's variance-partitioning approach to modeling the chromosomal determinants of a quantitative trait will be useful only for traits with high heritabilities or when fine-scale genetic maps can be employed. Goldgar's technique as a whole has promise, as it can be made relatively robust statistically, refined through some simple and intuitive extensions, and can be easily adapted to work with more complex sampling units. Further extensions of Goldgar's methods are proposed, and areas in need of additional research are discussed.

Genetic Markers↗
PubMed2006

Effects of sampling regime on the mean and variance of home range size estimates.

1. Although the home range is a fundamental ecological concept, there is considerable debate over how it is best measured. There is a substantial literature concerning the precision and accuracy of all commonly used home range estimation methods; however, there has been considerably less work concerning how estimates vary with sampling regime, and how this affects statistical inferences. 2. We propose a new procedure, based on a variance components analysis using generalized mixed effects models to examine how estimates vary with sampling regime. 3. To demonstrate the method we analyse data from one study of 32 individually marked roe deer and another study of 21 individually marked kestrels. We subsampled these data to simulate increasingly less intense sampling regimes, and compared the performance of two kernel density estimation (KDE) methods, of the minimum convex polygon (MCP) and of the bivariate ellipse methods. 4. Variation between individuals and study areas contributed most to the total variance in home range size. Contrary to recent concerns over reliability, both KDE methods were remarkably efficient, robust and unbiased: 10 fixes per month, if collected over a standardized number of days, were sufficient for accurate estimates of home range size. However, the commonly used 95% isopleth should be avoided; we recommend using isopleths between 90 and 50%. 5. Using the same number of fixes does not guarantee unbiased home range estimates: statistical inferences differ with the number of days sampled, even if using KDE methods. 6. The MCP method was highly inefficient and results were subject to considerable and unpredictable biases. The bivariate ellipse was not the most reliable method at low sample sizes. 7. We conclude that effort should be directed at marking more individuals monitored over long periods at the expense of the sampling rate per individual. Statistical results are reliable only if the whole sampling regime is standardized. We derive practical guidelines for field studies and data analysis.

Animals↗
PubMed2005

Simple nomograms to calculate sample size in diagnostic studies.

OBJECTIVES: To produce an easily understood and accessible tool for use by researchers in diagnostic studies. Diagnostic studies should have sample size calculations performed, but in practice, they are performed infrequently. This may be due to a reluctance on the part of researchers to use mathematical formulae. METHODS: Using a spreadsheet, we derived nomograms for calculating the number of patients required to determine the precision of a test's sensitivity or specificity. RESULTS: The nomograms could be easily used to determine the sensitivity and specificity of a test. CONCLUSIONS: In addition to being easy to use, the nomogram allows deduction of a missing parameter (number of patients, confidence intervals, prevalence, or sensitivity/specificity) if the other three are known. The nomogram can also be used retrospectively by the reader of published research as a rough estimating tool for sample size calculations.

Diagnostic Techniques and Procedures↗
PubMed2005

Experimental designs for multicomponent interventions among persons with multifactorial geriatric syndromes.

This paper discusses issues about the design of clinical trials to test multicomponent interventions for multifactorial health conditions, such as geriatric syndromes in which more than one risk factor is related to the outcome. The issues covered include: identification and selection of modifiable risk factors related to the outcome of interest, selection of intervention components to reduce the deleterious effects of the modifiable risk factors, assignment of components of the intervention, blinding, sample size requirements and estimation of component effects. Each of these issues is explored using examples from nine illustrative multicomponent intervention trials. Statistical and clinical concerns regarding the design of multicomponent interventions are addressed. We also propose elements of multicomponent interventions for multifactorial health conditions that should be reported in publications and areas where future research is needed.

Accidental Falls↗
PubMed1997

Jackknife bias reduction for polychotomous logistic regression.

Despite theoretical and empirical evidence that the usual MLEs can be misleading in finite samples and some evidence that bias reduced estimates are less biased and more efficient, they have not seen a wide application in practice. One can obtain bias reduced estimates by jackknife methods, with or without full iteration, or by use of higher order terms in a Taylor series expansion of the log-likelihood to approximate asymptotic bias. We provide details of these methods for polychotomous logistic regression with a nominal categorical response. We conducted a Monte Carlo comparison of the jackknife and Taylor series estimates in moderate sample sizes in a general logistic regression setting, to investigate dichotomous and trichotomous responses and a mixture of correlated and uncorrelated binary and normal covariates. We found an approximate two-step jackknife and the Taylor series methods useful when the ratio of the number of observations to the number of parameters is greater than 15, but we cannot recommend the two-step and the fully iterated jackknife estimates when this ratio is less than 20, especially when there are large effects, binary covariates, or multicollinearity in the covariates.

Bias↗
PubMed2004

Development of a new standard laboratory protocol for estimation of the field attenuation of hearing protection devices: sample size necessary to provide acceptable reproducibility.

The mandate of ASA Working Group S12/WG11 has been to develop "laboratory and/or field procedure(s) that yield useful estimates of field performance" of hearing protection devices (HPDs). A real-ear attenuation at threshold procedure was selected, devised, tested for one earmuff and three earplugs via an interlaboratory study involving five laboratories and 147 subjects, and incorporated into a new standard that was approved in 1997 [Royster et al., "Development of a new standard laboratory protocol for estimating the field attenuation of hearing protection devices. Part I. Research of Working Group 11, Accredited Standards Committee S 12, Noise," J. Acoust. Soc. Am. 99, 1506-1526; ANSI, S12.6-1997, "American National Standard method for measuring real-ear attenuation of hearing protectors" (American National Standards Institute, New York, 1997)]. The subject-fit methodology of ANSI S12.6-1997 relies upon listeners who are audiometrically proficient, but inexperienced in the use of HPDs. Whenever a new method is adopted, it is important to know the effects of variability on the power of the measurements. In evaluation of protector noise reduction determined by experimenter-fit, informed-user-fit, and subject-fit methods, interlaboratory reproducibility was found to be best for the subject-fit method. Formulas were derived for determining the minimum detectable difference between attenuation measurements and for determining the number of subjects necessary to achieve a selected level of precision. For a precision of 6 dB, the study found that the minimum number of subjects was 4 for the Bilsom UF-1 earmuff, 10 for the E.A.R Classic earplug, 31 for the Willson EP100 earplug, and 22 for the PlasMed V-51R earplug.

Auditory Threshold↗