Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Sampling Errors”

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 19 recordsLinked to original sources
PubMed2006

Comparison of analytical error and sampling error for contaminated soil.

Investigation of soil from contaminated sites requires several sample handling steps that, most likely, will induce uncertainties in the sample. The theory of sampling describes seven sampling errors that can be calculated, estimated or discussed in order to get an idea of the size of the sampling uncertainties. With the aim of comparing the size of the analytical error to the total sampling error, these seven errors were applied, estimated and discussed, to a case study of a contaminated site. The manageable errors were summarized, showing a range of three orders of magnitudes between the examples. The comparisons show that the quotient between the total sampling error and the analytical error is larger than 20 in most calculation examples. Exceptions were samples taken in hot spots, where some components of the total sampling error get small and the analytical error gets large in comparison. Low concentration of contaminant, small extracted sample size and large particles in the sample contribute to the extent of uncertainty.

Reproducibility of Results↗
PubMed1986

Tritiated thymidine labelling in vitro of human cancer of the breast: counting error and sampling error.

Tritiated thymidine labelling indices (TLIs) were determined on a number of primary cancers of the human breast. Twenty-two slides were chosen which demonstrated a wide range of TLIs and each was counted twice to assess 'counting error'. TLIs derived from successive counts of the same slide showed a coefficient of variance greater than 25% in half of the slides, but a significant difference between the two (P less than 0.05) in only 2 of 22 pairs. When TLIs derived from paired specimens taken from different sites in each of 22 tumours were compared, there was found to be a significant difference (P less than 0.05) in 17 of 22 pairs. This is 'sampling error' and is clearly a major source of inaccuracy when TLIs are derived from single small samples of heterogeneous tumours such as cancers of the human breast.

Breast Neoplasms↗
PubMed2005

A comment on sampling error in the standardized mean difference with unequal sample sizes: avoiding potential errors in meta-analytic and primary research.

The authors discuss potential confusion in conducting primary studies and meta-analyses on the basis of differences between groups. First, the authors show that a formula for the sampling error of the standardized mean difference (d) that is based on equal group sample sizes can produce substantially biased results if applied with markedly unequal group sizes. Second, the authors show that the same concerns are present when primary analyses or meta-analyses are conducted with point-biserial correlations, as the point-biserial correlation (r) is a transformation of d. Third, the authors examine the practice of correcting a point-biserial r for unequal sample sizes and note that such correction would also increase the sampling error of the corrected r. Correcting rs for unequal sample sizes, but using the standard formula for sampling error in uncorrected r, can result in bias. The authors offer a set of recommendations for conducting meta-analyses of group differences.

Humans↗
PubMed1986

Sampling errors in pH and blood gas analysis--an evaluation of three new arterial blood samplers.

We have tested the accuracy, acceptability and general performance of three recently-marketed samplers for arterial blood gas measurement (the Corning Arterial Blood Sampler, the Concord 'Pulsator' and the Sarstedt 'Monovette'). All three greatly reduce or eliminate the error of venous sampling, and the Corning and Sarstedt samplers eliminate the risk of dilution of the sample by excess heparin solution. A positive bias in pO2 measurement, more marked at higher levels, was demonstrated with the Concord and Sarstedt samplers, and the latter carry a slightly increased risk of cross-infection. None of the samplers completely overcame potential sampling errors.

Arteries↗
PubMed1992

A microcomputer program for evaluating sampling error: an application to stereological methods for electron microscopy.

In situations where there is a need to minimize sampling error or sample size, the coefficient of variation (CV) may be used to evaluate sampling error as a function of the number of observations or subjects in a sample. For example, CV is useful for estimating the minimum number of electron micrographs (Nmin) required to obtain a representative field sample for stereological analysis. To facilitate the determination of Nmin, we have written a program (COEFficient) for DOS microcomputers which calculates CVs. COEF assists the user in reducing error to that which solely reflects biological variability, thereby minimizing the time and cost of subsequent analyses.

Microcomputers↗
PubMed1996

An analysis of sampling errors for the Demographic Health Surveys.

"Sampling errors and design effects from 48 nationally representative surveys conducted under the Demographic and Health Surveys Program for a large number of variables concerning fertility, family planning, fertility intentions, child health and mortality etc. are analysed for the total sample, and for urban-rural domains, sub-national regions and various demographic and socio-economic subclasses.... At the country level, overall design effect (the ratio of actual to simple random sampling standard error) averaged over all variables and countries is around 1.5. Variation among countries is high, but less so than among variables. Urban-rural and regional differentials in design effects are small, and can be attributed to the fact that similar sample designs and cluster sizes were used across those domains within each country. Design effects for estimates over other subclasses are smaller, and tend towards 1.0 for small subclasses and differences, apart from the effect of sample weights which tends to persist undiminished across variables and subclasses." (SUMMARY IN FRE)

Child Welfare↗
PubMed2004

Quantitative analysis of spatial sampling error in the infant and adult electroencephalogram.

The purpose of this report was to determine the required number of electrodes to record the infant and adult electroencephalogram (EEG) with a specified amount of spatial sampling error. We first developed mathematical theory that governs the spatial sampling of EEG data distributed on a spherical approximation to the scalp. We then used a concentric sphere model of current flow in the head to simulate realistic EEG data. Quantitative spatial sampling error was calculated for the simulated EEG, with additive measurement noise, for 64, 128, and 256 electrodes equally spaced over the surface of the sphere corresponding to the coverage of the human scalp by commercially available "geodesic" electrode arrays. We found the sampling error for the infant to be larger than that for the adult. For example, a sampling error of less than 10% for the adult was obtained with a 64-electrode array but a 256-electrode array was needed for the infant to achieve the same level of error. With the addition of measurement noise, with power 10 times less than that of the EEG, the sampling error increased to 25% for both the infant and adult, for these numbers of electrodes. These results show that accurate measurement of the spatial properties of the infant EEG requires more electrodes than for the adult.

Adult↗
PubMed2001

Spectral noise due to sampling errors in Fourier-transform spectroscopy.

An assessment is made of the spectral noise in Fourier-transform spectroscopy caused by sampling errors in the interferogram acquisition. Numerical evaluations are performed in the case of the REFIR (radiation explorer in the far infrared) instrument developed for the measurement of the long-wavelength Earth emissions from satellite platforms. In this case the slow response of a room-temperature pyroelectric detector, the relatively short acquisition time, the broadband operation, and the wish for a relaxed requirement of the mirror drive accuracy make sampling error an important issue. Different sampling methods can be considered for reduction of the spectral noise induced by sampling errors. The effects of different sampling methods are quantified and discussed for the selection of the most-suitable option for this instrument. We find that only sampling methods that introduce some compensation (either analog or digital) of the frequency dependence of amplitude and phase components of the acquisition-system responsivity provide satisfactory results. In particular, the equal time sampling followed by a digital filter and numerical resampling has been examined minutely with a simulation model used to perform sensitivity tests of the main parameters that characterize the procedure.

Journal Article↗
PubMed1989

Variability in muscle fibre areas in whole human quadriceps muscle: how to reduce sampling errors in biopsy techniques.

A single biopsy is a poor estimator of the muscle fibre cross-sectional area (CSA) for a whole human muscle because of the large variability in the fibre area within a muscle. To determine how the sampling errors in biopsy techniques can be reduced, data on the CSA of type 1 and type 2 fibres obtained from cross-sections of whole vastus lateralis muscle of young men, have been analysed statistically. To obtain a good estimate of the mean fibre CSA in a biopsy, measuring all fibres in that biopsy gives the best result. To obtain a good estimate of the mean fibre CSA for a whole muscle, the number of biopsies has a much greater influence on the sampling error than the number of fibres measured in each biopsy, but the number of biopsies needed to obtain a given sampling error can vary by a factor of two. If the fibre CSA in three or more biopsies is measured, it is sufficient to measure only 25 fibres in each biopsy. If less than three biopsies are taken, there is no worthwhile reduction in sampling error when more than 100 fibres are measured. To determine the mean fibre CSA for a whole group of individuals, our preference is to maximize the number of individuals, and only take single biopsies. In conclusion, to determine the mean fibre CSA for this particular muscle with a certain precision, we suggest analysis of three biopsies, taken from different depths of the muscle, and measurement of 25 fibres in each biopsy.

Adolescent↗
PubMed2002

Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection.

OBJECTIVES: Needle liver biopsy has been shown to have a high rate of sampling error in patients with diffuse parenchymal liver diseases. In these cases, the sample of liver tissue does not reflect the true degree of inflammation, fibrosis, or cirrhosis, despite an adequate sample size. The aim of this study was to determine the rate and extent of sampling error in patients with chronic hepatitis C virus infection, and to assess the intraobserver variation with the commonly used scoring system proposed by Scheuer and modified by Batts and Ludwig. METHODS: A total of 124 patients with chronic hepatitis C virus infection underwent simultaneous laparoscopy-guided biopsies of the right and left hepatic lobes. Formalin-fixed paraffin-embedded sections were stained with hematoxylin and eosin and with trichrome. The slides were blindly coded and randomly divided among two hepatopathologists. Inflammation and fibrosis were scored according to the standard grading (inflammation) and staging (fibrosis) method based on the modified Scheuer system. Following the interpretation, the slides were uncoded to compare the results of the right and left lobes. Fifty of the samples were blindly resubmitted to each of the pathologists to determine the intraobserver variation. RESULTS: Thirty of 124 patients (24.2%) had a difference of at least one grade, and 41 of 124 patients (33.1%) had a difference of at least one stage between the right and left lobes. In 18 patients (14.5%), interpretation of cirrhosis was given in one lobe, whereas stage 3 fibrosis was given in the other. A difference of two stages or two grades was found in only three (2.4%) and two (1.6%) patients, respectively. Of the 50 samples that were examined twice, the grading by each pathologist on the second examination differed from the first examination in 0% and 4%, and the staging differed in 6% and 10%, respectively. All observed variations were of one grade or one stage. CONCLUSIONS: Liver biopsy samples taken from the right and left hepatic lobes differed in histological grading and staging in a large proportion of chronic hepatitis C virus patients; however, differences of more than one stage or grade were uncommon. A sampling error may have led to underdiagnosis of cirrhosis in 14.5% of the patients. These differences could not be attributed to intraobserver variation, which appeared to be low.

Adult↗
PubMed1978

Subhourly variability of circulating third trimester maternal steroid concentrations as a source of sampling error.

Seven plasma samples from five normal third trimester pregnant women, drawn every 5 and 15 min at similar times on 2 days, 2 days apart, were measured in quadruplicate for estradiol (E2), estriol, progesterone, 16 alpha-hydroxyprogesterone, and 17 alpha-hydroxyprogesterone (17P). The mean of the 22 samples obtained from each subject for each steroid was used as a reference mean. Individual determinations were converted to percentages of the reference mean which was normalized to 0%. All pregnancies were uncomplicated, and all mean values were within the normal range for gestational age. Variability about the reference mean for single samples, however, range from a low of -80% to a high of 150%. The single sample, 95% confidence intervals for individual steroids pooled from the five subjects (110 determinations) range from +/- 36% (E2) to +/- 60% (17P). Mean percentage coefficients of variation between 5-min and 15-min sampling sequences were compared by analysis of variance. There is no significant difference between the mean percentage coefficients of variation of a 5-min as opposed to a 15-min sampling sequence for any of the hormones measured. The 95% confidence interval width around the reference mean is a function of the number of samples obtained. Because the 95% confidence interval width from 110 measurements decreases approximately as 1/ square root n with increasing sampling size, the decrement progressively diminishes. For E2, the least variable steroid, a one-sample 95% confidence interval width of +/- 36% decreases to approximately +/- 18% with four samples or approximately +/- 12% with nine samples. For 17P, the most variable steroid, a one-sample 95% confidence interval width of +/- 60% decreases to approximately +/- 30% with four samples or approximately +/- 20% with nine samples. Multiple sampling with plasma pooling is required for the accurate study of steroid concentrations in individual subjects in late pregnancy.

17-alpha-Hydroxyprogesterone↗
PubMed1985

Analysis of sampling errors in biopsy techniques using data from whole muscle cross sections.

Because of the large variability in the proportion of fiber types within a whole muscle, a single biopsy is a poor estimator of the fiber type proportion for a whole muscle. Data on the proportions of type I and II fibers, obtained from cross sections of whole human muscles (vastus lateralis) from young male individuals, have therefore been analyzed statistically in order to determine the sampling errors involved in muscle biopsy techniques. For the purpose of obtaining a good estimate of the fiber type proportion in a whole biopsy, counting all fibers is of great benefit compared with counting only half of the fiber number. The required number of biopsies to obtain a given sampling error of the mean proportion of fiber types in the whole muscle can vary by a factor of six. If less than three biopsies are taken from a muscle, there is a substantial reduction in sampling error taking biopsies with at least 600 fibers. For more than three biopsies there is a small gain in sampling greater than 150 fibers. The precision of the estimate of the mean proportion of fiber types for a group is increased with the number of biopsies per individual and number of individuals. In conclusion, for the muscle in this study, complete counting of three biopsies, each greater than 150 fibers, sampled from different depths of the muscle is recommended.

Adolescent↗
PubMed1997

The effect of sampling error and measurement error and its correlation on the estimation of multi-locus fixed-bin VNTR RFLP genotype probabilities.

Bootstrapping was used to examine the effect of sampling error and measurement error and its correlation on fixed-bin genotype probabilities. Bootstrap confidence intervals (Cls) were made relative to the point estimate using the log of the inverse of the probabilities. From databases of 200-250 genotypes, sampling error alone yielded median relative 95% CIs of from one order of magnitude out of five for one locus to one out of ten for four loci. Measurement error of the test genotype fragments increased the latter to about one order of magnitude out of eight. Database measurement error and its correlation had only a slight effect on multi-locus probability uncertainty. Together, these uncertainties are several orders of magnitude greater than error due to population substructuring of a race by its major component ethnic groups.

Confidence Intervals↗
PubMed1994

Estimation of blood sampling errors resulting from metabolism and solute exchange between plasma and formed elements.

The origin and magnitude of potential errors in whole-blood sampling are predicted on the basis of a mathematical model. The model describes the kinetics of solute metabolism, breakdown, and interphase distribution (i.e., partitioning and exchange between formed elements and plasma) within a blood sample during sample withdrawal and storage. The model is applied to the determination of the integral over time of solute concentration in the plasma (area-under-the-curve, or AUC) from a sample withdrawn through an arterial or venous catheter. Errors in AUC determination can be substantial and are strongly dependent on the duration of sampling (T), the rate constants for solute degradation processes, the rate constant for solute exchange between the formed elements and the plasma (ke), and the equilibrium ratio for distribution of the solute between formed elements and plasma (R). When the value of the dimensionless group keT/R is small, little solute exchanges between plasma water and formed elements before the two phases of the blood are separated. When keT/R is large, the solute distribution is close to equilibrium at all times. In these two keT/R limits, the contribution of solute redistribution to sampling error is small. Sizable errors resulting from redistribution are associated with intermediate values of keT/R, even in the absence of metabolism and despite rapid separation of the phases at the end of the withdrawal period. Chemical conversion within either of the blood phases introduces additional sampling error under most circumstances.

Blood Cells↗
PubMed1998

[Analysis of variance of bacterial counts in milk. 1. Characterization of total variance and the components of variance random sampling error, methodologic error and variation between parallel errors during storage].

In contrast to the prevailing automatized chemical analytical methods, classical microbiological techniques are linked with considerable material- and human-dependent sources of errors. These effects must be objectively considered for assessing the reliability and representativeness of a test result. As an example for error analysis, the deviation of bacterial counts and the influence of the time of testing, bacterial species involved (total bacterial count, coliform count) and the detection method used (pour-/spread-plate) were determined in a repeated testing of parallel samples of pasteurized (stored for 8 days at 10 degrees C) and raw (stored for 3 days at 6 degrees C) milk. Separate characterization of deviation components, namely, unavoidable random sampling error as well as methodical error and variation between parallel samples, was made possible by means of a test design where variance analysis was applied. Based on the results of the study, the following conclusions can be drawn: 1. Immediately after filling, the total count deviation in milk mainly followed the POISSON-distribution model and allowed a reliable hygiene evaluation of lots even with few samples. Subsequently, regardless of the examination procedure used, the setting up of parallel dilution series can be disregarded. 2. With increasing storage period, bacterial multiplication especially of psychrotrophs leads to unpredictable changes in the bacterial profile and density. With the increase in errors between samples, it is common to find packages which have acceptable microbiological quality but are already spoiled by the time of the expiry date labeled. As a consequence, a uniform acceptance or rejection of the batch is seldom possible. 3. Because the contamination level of coliforms in certified raw milk mostly lies near the detection limit, coliform counts with high relative deviation are expected to be found in milk directly after filling. Since no bacterial multiplication takes place during storage, then error between samples always predominates the total variation. 4. The present results cannot be simply applied to other selective enumerations of microorganisms. Yet, a non-homogenous distribution should always be expected at microbial counts close to the detection limit. Technical errors arising from clustering as well as eugonic growth can additionally hamper the counting of colonies of such microorganisms. Effects of these observations in the decision-making process will be dealt with in the second communication.

Analysis of Variance↗
PubMed2004

On sampling and sampling errors in histomorphometry of peripheral nerve fibers.

Histomorphometrical assessment of regenerated peripheral nerves is a very common goal of many studies in experimental microsurgery. In this paper, the main critical issues in nerve fiber sampling for quantitative morphological assessment are addressed. The equal opportunity rule, i.e., the basic paradigm of random sampling, is described, together with an explanation of how sampling errors, in the selection of histologic fields and of the nerve fibers inside them, can produce a bias in quantitative estimates. Finally, some practical suggestions on how to cope with the most common sampling errors are provided, in order to help researchers obtain reliable histomorphometrical data on peripheral nerve fibers.

Humans↗