ASP
Statistical Software For Business, Economics, And The Social Sciences |
|
ReliabilityON THIS PAGE Reliability AnalysisThere are 13 reliability analysis options available on the RELIABILITY ANALYSIS OPTIONS menu:
Item And Scale AnalysisThe I. Item And Scale Statistics option on the RELIABILITY ANALYSIS OPTIONS menu provides the number of items in the scale and the number of cases along with the scale mean, variance, and standard deviation. In addition, it provides the grand mean, Cornbach’s alpha, standardized alpha, interclass correlation, and the Spearman Brown predictor. This option also provides the mean, variance, standard deviation, covariance, correlation of the given item with the sum of the other items, and the reliability index of each item in the scale. It also provides the mean, variance, standard deviation, maximum, minimum, range, and maximum over the minimum of each of the proceeding statistics. Item AnalysisThere are four item analysis options on the RELIABILITY ANALYSIS OPTIONS menu:
Variance/Covariance MatrixThe V. Variance/Covariance Matrix option on the RELIABILITY ANALYSIS OPTIONS menu provides the variance/covariance matrix of the items in the scale. Affect Of Deleting ItemThe D. Affect Of Deleting Item option on the RELIABILITY ANALYSIS OPTIONS menu shows how the mean, variance, and alpha of the scale will change if a given item is excluded from the scale. Quartile BreakdownThe Q. Quartile Breakdown Mean option on the RELIABILITY ANALYSIS OPTIONS menu provides a quartile breakdown of the mean of each item in the scale. The B . Quartile Breakdown Std Dev option on the RELIABILITY ANALYSIS OPTIONS menu provides a quartile breakdown of the standard deviation of each item in the scale. Scale AnalysisThere are seven scale analysis options on the RELIABILITY ANALYSIS OPTIONS menu:
Analysis Of VarianceThe A. Analysis Of Variance option on the RELIABILITY ANALYSIS OPTIONS menu provides the analysis of variance table for the reliability model being examined. If the items in the scale are non-dichotomous this includes the between measures and nonadditivity F statistics along with the Tukey additivity power. If the items in the scale are dichotomous only the between measures F statistic is included. Split ModelThe S. Split Model option on the RELIABILITY ANALYSIS OPTIONS menu divides the items in the scale into two groups, the first half of the items in the first group and the second half of the items in the second group. (If there is an odd number of items the first group will have one more item than the second.) The correlation between the two groups is then calculated along with the Spearman Brown coefficient and the Guttman split half coefficient. The various scale statistics provided by option I are also provided for each half of the scale. Guttman ModelThe G. Guttman Model option on the RELIABILITY ANALYSIS OPTIONS menu provides the values for L1 through L6 of the Guttman Model. Strict Parallel ModelThe P. Parallel Model option on the RELIABILITY ANALYSIS OPTIONS menu provides the chi square test for goodness of fit for the parallel model along with its degrees of freedom, p-value, and all of the statistics necessary to generate this test. The T. Strict Parallel Model option on the RELIABILITY ANALYSIS OPTIONS menu provides the chi square test for goodness of fit for the strict parallel model along with its degrees of freedom, p-value, and all of the statistics necessary to generate this test. Concordance/Friedman’s TestThe C. Concordance/Friedman’s Test option on the RELIABILITY ANALYSIS OPTIONS menu is only available when the scale items are dichotomous. In this case Friedman’s chi square (along with its degrees of freedom, and p-value) and Kendall’s coefficient of concordance are provided along with all of the statistics necessary to generate these two items. Hotelling’s T Squared TestThe H. Hotelling’s T Squared Test option on the RELIABILITY ANALYSIS OPTIONS menu provides Hotelling’s T squared statistic along with its F statistic, degrees of freedom, and p-value. |
|
If you have
any problems, comments, or suggestions regarding this web site,
please notify the webmaster at
webmaster@dmcSoftware.com. |