Variance and Correlation Coefficients

Berger, Richard A.; Sweney, Arthur B.

doi:10.1080/10671188.1965.10614705

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As we deducted from Figure , such drops in correlation appear at similar timescales when precipitation events are not synchronized, a phenomenon that should be present in Yanco because of irrigation. Also taking into account that the 32 day scale has small temporal variance, similar to that of the 2 day scale for this data set (not shown here), the signal‐to‐noise ratio might be quite low and induce low correlation (as demonstrated by Berger & Sweney, ; Goodwin & Leech, ).…”

Section: Resultsmentioning

confidence: 62%

Multi‐Timescale Analysis of the Spatial Representativeness of In Situ Soil Moisture Data within Satellite Footprints

et al. 2018

View full text Add to dashboard Cite

We conduct a novel comprehensive investigation that seeks to prove the connection between spatial scales and timescales in surface soil moisture (SM) within the satellite footprint (~50 km). Modeled and measured point series at Yanco and Little Washita in situ networks are first decomposed into anomalies at timescales ranging from 0.5 to 128 days, using wavelet transforms. Then, their degree of spatial representativeness is evaluated on a per‐timescale basis by comparison to large spatial scale data sets (the in situ spatial average, SMOS, AMSR2, and ECMWF). Four methods are used for this: temporal stability analysis (TStab), triple collocation (TC), percentage of correlated areas (CArea), and a new proposed approach that uses wavelet‐based correlations (WCor). We found that the mean of the spatial representativeness values tends to increase with the timescale but so does their dispersion. Locations exhibit poor spatial representativeness at scales below 4 days, while either very good or poor representativeness at seasonal scales. Regarding the methods, TStab cannot be applied to the anomaly series due to their multiple zero‐crossings, and TC is suitable for week and month scales but not for other scales where data set cross‐correlations are found low. In contrast, WCor and CArea give consistent results at all timescales. WCor is less sensitive to the spatial sampling density, so it is a robust method that can be applied to sparse networks (one station per footprint). These results are promising to improve the validation and downscaling of satellite SM series and the optimization of SM networks.

show abstract

Section: Resultsmentioning

confidence: 62%

Multi‐Timescale Analysis of the Spatial Representativeness of In Situ Soil Moisture Data within Satellite Footprints

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Some authors (2,7,11) have suggested that the choice of the criterion score should be based on which score will produce the highest correlation coefficient. The present author suggests that the experimenter should attempt to approximate the true score as closely as possible, and that this should not be sacrificed to enable the experimenter to report a higher correlation coefficient regardless of whether reliability, specificity, or any other relationships are being estimated.…”

Section: Discussionmentioning

confidence: 99%

“…Research Quarterly articles by Henry (3,4,7), McCraw (10), Whitley and Smith (11), Berger and Sweney (2) and Kroll (9) have all discussed this question. Berger and Sweney, and McCraw have recommended the utilization of a best score, whereas, the other authors have favored the average score.…”

Section: Within-subject Variation Measurement Error and Selection Omentioning

confidence: 96%

Within-Subject Variation, Measurement Error, and Selection of a Criterion Score

Hetherington

1973

Research Quarterly. American Association for Health, Physical E

View full text Add to dashboard Cite

“…Frequently, investigators choose the subject's best score of several trials as his representative strength score. Though this probably evolved through convenience, coefficients of reliability have been found (1,8,15,22) to change if individual best scores were correlated rather than average scores. Other investigators (7,12,14) are in disagreement.…”

mentioning

confidence: 97%

Reliability Estimation in the Measurement of Strength

Alderman

Banfield

1969

Research Quarterly. American Association for Health, Physical E

View full text Add to dashboard Cite

This study was undertaken to establish the reliability of six cable tension and two dynamometric measures of muscular strength using the Hettinger Strength Chair (modified by Howell). Thirty-two college freshman males served as subjects. The strength chair was designed to prevent shoulder rotation, provide body stability, and an objective reduplication of test posture. Subjects were given three trials on each of eight measures on four different occasions. Reliability coefficients of the eight isometric measures ranged from moderately high (.74) to high (.98). A random administration of test items did not result in significant differences in reliability when compared to a standard order. Neither did the use of best scores rather than average scores result in any changes in reliability.DURING THE PAST DECADE a considerable amount of interest has arisen with respect to the basic level of muscular strength required for the efficient and effective performance of daily tasks and activities. Though extensive information exists on the nature of this basic level, valid and reliable conclusions have not been forthcoming. This indecisiveness has been generally predicated on both the weaknesses in the method of cable tensiometry measurement and in the various statistical methods employed in determining the reliability of data obtained.

show abstract

Variance and Correlation Coefficients

Cited by 3 publications

References 0 publications

Multi‐Timescale Analysis of the Spatial Representativeness of In Situ Soil Moisture Data within Satellite Footprints

Multi‐Timescale Analysis of the Spatial Representativeness of In Situ Soil Moisture Data within Satellite Footprints

Within-Subject Variation, Measurement Error, and Selection of a Criterion Score

Reliability Estimation in the Measurement of Strength

Contact Info

Product

Resources

About