Showing papers on "Sampling (statistics) published in 1975"

An Investigation of the Parallel Analysis Criterion for Determining the Number of Common Factors

Abstract: Then correlation matrices based upon real and random data with squared multiple correlations in the diagonals are factored, the hypothesis is that the point at which the curves of the latent roots cross indicates the number of common factors. Sampling studies confirm the hypothesis when the common factor model provides a good fit to the data. When small overlapping, nonrandom factors are introduced, the expected value of the number of common factors can still be the number of major factor in the population when the nonrandom "noise" is small compared to sampling error "noise." This criterion for the number of common factors, furthermore, is more accurate than the method of maximum likelihood.

An evaluation of time-sample measures of behavior.

Abstract: Continuous and time-sample measures of the in-seat behavior of a secretary were obtained. Measurement error, i.e., the extent to which the sample measures deviated from the continuous measure, was a function of the frequency of the sample measurements and the criterion used to score an example of the behavior. If the behavior had to be exhibited throughout the observational interval (whole-interval time sampling), there was a consistent underestimate of the continuous measure. If the behavior had to be exhibited only briefly within the observational interval (partial-interval time sampling), there was a consistent overestimate of the continuous measure. And, if the behavior had to be exhibited at the end of the observational interval (momentary time sampling), overestimations and underestimations of the continuous measure occurred about equally often. As expected, the more frequently the sample measures were made the closer was the agreement between the sample and continuous measures. Two conclusions concerning measurement error in interval time sampling were made. The first was that the error will be a function of the mean time per response. The second is that this error will not be consistent across experimental conditions.

Copper in surface waters south of New Zealand

Abstract: THE sampling of the transition metals, including copper, in the ocean has so far produced unsatisfactory results. Reported measurements show quite a large scatter within a given suite of data and significant differences between investigators1–3. Published profiles in general look quite unlike any of the ‘accepted’ distributions for other dissolved species outlined later nor are there any close similarities between the transition elements themselves. Either they have a unique geochemistry or the data are to some degree invalid. For copper, contamination during sampling has been suggested4 as one source of error, and we have tried to overcome this by sampling surface waters in regions of marked horizontal gradients in chemical properties.

On Estimating the Aboveground Weights of Tree Stands

Abstract: Simulated sampling of nine tree stands was used to compare estimated stand component weights with known weights based on complete harvest. On average, estimates based on regressions of logarithm weight on either logarithm tree diameter or on logarithm (diameter) squared x height overestimated weights of stems, branches, and leaves by approximately 3% when the inherent bias of such regressions was accounted for. Ninety-five percent confidence intervals based on random sampling encompassed the measured stand weights 79%-100% of the time. Similar confidence intervals using stratified random sampling based on tree diameter encompassed the measured stand weights only 44%-98% of the time. The small average bias in estimates of stand weights using logarithmic regressions is of minor importance compared with the variation in estimates among replicated samples. 21 references.

Stratigraphic biases in the analysis of taxonomic survivorship

Abstract: Taxonomic survivorship curves may reflect stratigraphic as well as biologic factors. The approximately lognormal distribution of lengths of Phlanerozoic time intervals produces an error in the estimation of taxonomic durations that is also lognolrmally distributed. As demon- strated by several simulated examples, this error may cause concave taxonomic survivorship curves to appear linear, especially if the maximum durations involved are relatively short. The error of estimation also makes highly concave taxonomic survivorship curves virtually unrecog- nizable. Incomplete sampling of the fossil record, on the other hand, may not be a serious problem in survivorship analysis. Simulated paleontological sampling employing a simple model suggests that survivorship curves tend to retain their original shapes even when as few as 20% of the taxa have been discovered. However, concave taxonomic survivorship curves tend to lose their concavity as efficiency of sampling declines.

A new method of sequential sampling to classify populations relative to a critical density

Abstract: A method of sequential sampling for grading population level in relation to a critical density is proposed. The method is based on the\(\mathop m\limits^* - m\) relationship and can be used without restrictions on the distribution patterns. The formulae for simple random sampling as well as for two-stage sampling are given.

Fluid sampling vessel

Abstract: A sampling vessel, adpated particularly for use in geothermal bores, constructed principally of stainless steel and opened at depth without ancillary means such as messengers by causing an inertial mechanism to break a frangible seal.

Sampling and Sample Size in Ecological Analyses of Fossil Mammals

Abstract: Analysis of several thousand mammalian fossils from late Pleistocene sediments in California provide data on sampling in mammalian paleoecology. Recovery of bones and teeth from the screenwashed bulk sediment sample residue is considered nearly total. Neither surface collecting alone, nor small bulk samples provide satisfactory quantitative data on original community structure or postmortem alterations in community organization. Minimum sample size for the analysis of diversity is discussed. Diversity and size-trophic ratios of the total identifiable mammalian component of this fauna (N = 1222) are similar to those expected in living communities, and therefore suggest adequate sampling, and minimally biased samples.

An evaluation of the calibrated weight-estimate method for measuring production in annual vegetation.

Abstract: Highlight: A double sampling technique, of visual weight estimates calibrated by harvesting, was applied to the measurement of biomass and production curves in grazed and ungrazed semiarid annual grassland. Good levels of accuracy can be achieved in such vegetation with a time expenditure significantly lower than by harvesting only. Some methodological problems were encountered and solved by modifications of the method. In some conditions the method can be used for estimating animal intake. In a study of plant and animal production on semiarid annual grasslands at Migda in Israel (Tadmor et al., 1974), frequent, extensive, and accurate measurements of plant biomass were needed. To determine the seasonal growth curve and the differences in it between plots with different grazing and fertilization treatments, plot means had to be estimated with an accuracy of lo-15% every 2 weeks. The annual vegetation sampled showed great local variability in yield and composition, particularly in early growth stages, when the coefficient of variation of biomass in 25 X 25-cm squares within a field may be 100-l 50%. Thus sampling by harvesting alone to the required accuracy would have imposed a prohibitive work load. Therefore, an attempt was made to apply the double sampling technique of visual weight estimates calibrated by harvesting of a subsample to the annual vegetation concerned. This method is based on the observation that a trained estimator can achieve consistent estimates of biomass (Pechanec and Pickford, 1937), which can be converted to true values by using a calibration curve obtained on the same day (Wilm et al., 1949; Brown, 1959; Morley et al., 1969). The statistical aspects of the method have been discussed in Range Research, Basic Problems and Techniques (National Research Council, 1962). The extensive application of this method to grazed and ungrazed annual vegetation allows a thorough evaluation of its performance in this vegetation type, in terms of accuracy and

Sampling Characteristics of Kelley's ε and Hays' ω:

Abstract: Statistics used to estimate the population correlation ratio were reviewed and evaluated. The sampling distributions of Kelley's e2 and Hays' ω2 were studied empirically by computer simulation within the context of a three level one-way fixed effects analysis of variance design. These statistics were found to have rather large standard errors when small samples were used. As with other correlation indices, large samples are recommended for accuracy of estimation. Both e2 and ω2 were found to be negligibly biased. Heterogeneity of variances had negligible effects on the estimates under conditions of proportional representativeness of sample sizes with respect to their population counterparts, but combinations of heterogeneity of variance and unrepresentative sample sizes yielded especially poor estimates.

Computer system performance indicator

Abstract: Computer monitor instructions are placed in appropriate locations in a routine to be monitored. Upon their execution, a decoder appropriately activates or de-activates circuitry, which develops time-related performance parameters. These time-related parameters are produced by a first counter which counts to one second and then resets a second, sampling counter. The sampling counter is stepped by clock pulses gated to it in response to a decoder activation signal. Each second the sampling counters contents are gated to a register to control the height of a bar graph display. Alternatively, the sampling counter contents may be summed over a number of intervals and divided by that number of intervals to produce an average count for controlling the bar graph display. Lastly, the number of decoder activations may be counted over the one second interval to provide a rate indication.

Computer Analysis of Topography: A Comparison of Terrain Storage Methods

Abstract: Topographic information can be digitized in several ways. Sampling may be surface-random or surface-specific. Surface-random sampling includes grids, contours, and randomly-located points. In this ...

The Effect of Bias on Estimators of Relative Risk for Pair-Matched and Stratified Samples

Abstract: Monte Carlo methods are employed to compare the effectiveness of pair-matched and independent stratified samples for estimating relative risk in the presence of bias. Three approximations to the maximum likelihood estimator for stratified samples suggested by Woolf, Mantel and Haenszel and Birch, respectively [25, 17, 2], are also compared. The sampling model is modified to approximate the practical choices for a researcher and to allow for the loss of unmatchable sample units. The mean square error is always largest for the matched-pairs estimator, while of the stratified estimators, Woolf's consistently produces the smallest MSE, equaled only by Birch's when the samples are equal.

Estimating land use characteristics for hydrologic models

Abstract: Many hydrologic models, including both linked process simulation models and unit hydrograph models, require estimates of land use characteristics (e.g., percentage of impervious area) for calibration. Past investigations have demonstrated that reliable estimates of land use can be obtained by aerial photograph sampling analyses. Even when aerial photographs are available, however, it is an expensive and time-consuming process to measure land use characteristics. Prediction equations that are derived from data characterized by high variability in the land use characteristics appear to be a reliable alternative to aerial photograph sampling analyses. In addition to being less expensive and time consuming the prediction equation approach can be used in generating land use estimates for future land use configurations. The prediction equations derived herein require demographic characteristics (e.g., population density and housing density) that can be obtained from census summaries or projections compiled by planning agencies and thus could be used with hydrologic models to estimate the impact of future development alternatives.

The Collection of Sensitive Information Using a Two-Stage, Randomized Response Model:

Abstract: This article extends the randomized response sampling design to find the intensity of positive action on a sensitive topic among those who have taken a positive action. Sampling properties of the r...

Telephone circuit monitoring system

Abstract: A telephone circuit monitoring system includes sampling means for repeatedly sampling the bistate and linear (a-c and d-c) control signals concurrently on a plurality of trunk circuits. When any of the circuits go off-hook, a central control initiates the sampling operation and repeatedly directs the sampling of the linear signals and bistate signals of each of a plurality of off-hook circuits until the monitoring purpose has been served.