Analytical Solution for the Optimal Addition of an Item to a Composite of Scores for Maximum Reliability
This paper presents a derivation of the optimal weight to be assigned for an item so that it maximally increases the reliability of the aggregate. This aggregate is the best estimate of the underlying true repeating pattern. The approach differs from previous solutions in being analytical, based on the Signal to Noise Ratio (SNR) instead of the reliability itself, and the ability to visually inform the researcher about the relevance of the weighting strategy and the gains produced in the SNR. Optimal weighting of repetitive phenomena is a bonus not only in the behavioral sciences, but also in many engineering fields. Its uses may include the selection or discarding of raters, judges, repetitions, or epochs, depending on the field.
KeywordsReliability Signal-to-Noise Ratio Composites Ensemble Averages
This work was partially supported by an Alexander von Humboldt Foundation Fellowship granted to one of the authors (Ref 3.2-1164728-CUB-GF-E).
- 5.Kotsakis, C., Tziavos, I.N.: Parametric versus non-parametric methods for optimal weighted averaging of noisy data sets. In: Sansò, F. (ed.) A Window on the Future of Geodesy. IAG SYMPOSIA, vol. 128, pp. 434–439. Springer, Heidelberg (2005). https://doi.org/10.1007/3-540-27432-4_74CrossRefGoogle Scholar
- 6.Hashimoto, K., et al.: A novel signal-averaged electrocardiogram and an ambulatory-based signal-averaged electrocardiogram show strong correlations with conventional signal-averaged electrocardiogram in healthy subjects: a validation study. J. Electrocardiol. 51(6), 1145–1152 (2018)CrossRefGoogle Scholar
- 14.Sočan, G.: Assessment of reliability when test items are not essentially t-equivalent. Dev. Surv. Methodol. 15, 23–35 (2000)Google Scholar
- 15.Chang, S.-W.: Choice of weighting scheme in forming the composite. Bull. Educ. Psychol. 40, 489–510 (2009)Google Scholar
- 20.Telle, A., Vary, P.: A novel approach for impulse response measurements in environments with time-varying noise. In: Proceedings of the 20th International Congress on Acoustics, ICA 2010, Sydney, Australia, 23–27 August 2010, pp. 1–5, July 2010Google Scholar
- 21.Pander, T., Przybyla, T., Czabanski, R.: An Application of the LP-norm in robust weighted averaging of biomedical signals. J. Med. Inform. Technol. 22(2), 1–8 (2013)Google Scholar