Downloads & Free Reading Options - Results

Prior to conducting a statistical analysis, sufficient data screening methods should be used for all research variables to identify miscoded, missing, or otherwise messy data. The primary purpose of these exercises was to demonstrate the role of data screening techniques and their potential to improve the performance of statistical methods. A heuristic data set was used to make the discussion more concrete, and the Statistical Package for the Social Sciences (SPSS) was used to screen the data. Overall, cleaning raw data by determining normality and linearity problems, outlier influences, and missing value presence proved to increase the R squared values if only by very small increments. One of the most interesting findings in this exercise was the performance of the regression models when outliers were taken into consideration, without respect to any additional data cleaning procedures. These screening procedures, if used properly, assist the researcher in optimizing data so that the analysis procedure will produce the most accurate and efficient estimates. Two appendixes contain SPSS syntax for the analyses. (Contains 1 table, 9 figures, and 15 references.) (Author/SLD)

Prior to conducting a statistical analysis, sufficient data screening methods should be used for all research variables to identify miscoded, missing, or otherwise messy data. The primary purpose of these exercises was to demonstrate the role of data screening techniques and their potential to improve the performance of statistical methods. A heuristic data set was used to make the discussion more concrete, and the Statistical Package for the Social Sciences (SPSS) was used to screen the data. Overall, cleaning raw data by determining normality and linearity problems, outlier influences, and missing value presence proved to increase the R squared values if only by very small increments. One of the most interesting findings in this exercise was the performance of the regression models when outliers were taken into consideration, without respect to any additional data cleaning procedures. These screening procedures, if used properly, assist the researcher in optimizing data so that the analysis procedure will produce the most accurate and efficient estimates. Two appendixes contain SPSS syntax for the analyses. (Contains 1 table, 9 figures, and 15 references.) (Author/SLD)

We propose an adaptive denoising scheme for poorly modeled non-Gaussian features in the gravitational wave interferometric data. Preliminary tests on real data show encouraging results.