Brightstat

Brightstat is a software as a service (SaaS) statistical package offering basic statiscal analysis, data processing and manipulation as well as the creation and manipulation of sophisticated interactive graphs. Brightstat was developed by Daniel Stricker and was first published in 2008 as a Adobe(TM) Flash application. A complete redesign was done und Brighstat was relaunched as a modern SaaS website running on all devices (desktop, tablets and mobiles) in April 2015.

Statistical tests

Brighstat offers the following statistical computations and tests:

• Descriptive analysis including: Arithmetic Mean, Geometric Mean, Harmonic Mean, Mode, Median, Minimum, Maximum, Sum, Standard Deviation, Variance, Standard Error of the Mean, Kurtosis and Skewness.
• Frequencies Percent and cumulative percent. (Percentiles and Tukey Hinges)

Non-parametric Tests:

• Binomial test
• kxl Chi-Square including likelihood Ratio, Phi, Cramer's V, Kendall's tau-b and tau-c, Cohen's Kappa.
• McNemar's test including McNemar exact, likelihood Ratio, Phi, Cramer's V, Symmetric Measures.
• Fisher's Exact Test
• Mann-Whitney U test for comparison of two independent samples.
• Kruskal-Wallis test for comparison of more than two independent samples. Including paired comparisons for post-hoc analysis.
• Wilcoxon and Sign-Test for comparison of two dependent samples.
• Friedman test for comparison of more than two dependent samples. Including Conover Statistic and paired comparison for post-hoc analysis. Including Kendall W
• Kolmogorov-Smirnov one sample test for testing against Uniform, Normal and Exponential distribution. Lilliefors correction for test against normality.
• Kolmogorov-Smirnov two sample test

Parametric Tests:

• One-Sample t-test for testing sample mean against mean of population.
• Two independent Samples t-test for mean comparison of two independent samples. Including Levene test for homogeneity of variances.
• Two dependent Samples t-test for mean comparison of two dependent sample. Including paired correlations.
• One-Way ANOVA including post hoc test LSD, Tukey HSD, Scheffe, Sidak and Bonferroni for equal variances assumed and Games-Howell GH, Tamhane T2, for equal variances not assumed.
• Repeated One-Way ANOVA including estimation of Epsilon (LowerBound, Greenhouse-Geisser, Huynh-Feldt) including post hoc test LSD, Tukey HSD, Scheffe, Sidak and Bonferroni for observed means and LSD, Sidak and Bonferroni for estimated marginal means.
• ANOVA, univariate analysis of variances for fixed factors, uncluding Type III sum of squares, Levene test for homogeneity of variances, eta squared and partial eta squared, including post hoc test LSD, Tukey HSD, Scheffe, Sidak and Bonferroni for equal variances assumed and Games-Howell GH, Tamhane T2, for equal variances not assumed.
• ANOVA Split-plot including Test for Sphericity, Levene test for homogeneity of variances, including post hoc test LSD, Tukey HSD, Scheffe, Sidak and Bonferroni for observed means and LSD, Sidak and Bonferroni for estimated marginal means, up to 1 within and n between factors,
• Multivariate ANOVA (MANOVA) including Wilks Lambda, Pillai's trace and Hotelling's trace
• Pearson bivariate correlations.
• Spearman rank correlations.
• Multiple Linear Regression, including Correlationmatrix, Model-Fit, ANOVA-Table, Parameter Estimates, Part and Partial Correlations, Collinearity (Tolerance and Variance Inflation Factor), Change Statistics. Methods: Enter, Forward, Backward and Step.
• Reliability Analysis, including Correlationmatrix, Covariancematrix, ANOVA-F-Table, Item statistics, Item Total-statistics, Scale-statistics, Summaries, Hotelling T-Squared, Tukey test for nonadditivity, Intraclass correlations. Models: Alpha, split-half, Guttman, parallel, strict-parallel.
• One-Sample Variance Test
• k-Sample Variance Test (Levene's Test for homogeneity of variances)

Graph types

• Line graph: Single and multiple line graphs for summaries of group of cases or summaries of variables.
• Bar graph: Simple and clustered bar Graphs for summaries of group of cases or summaries of variables.
• Area graph: Simple and clustered area graphs for summaries of group of cases or summaries of variables.
• Scatterplot with data clustering and the following fit-line functions: linear, logarithmic, inverse, quadratic, cubic, growth, exponential, compound, power, logistic
• Histogram: definable category width, normal distribution plot
• Box plot

References

Stricker, D. (2008). "Brightstat.com - free statistics online". Computer Methods and Programs in Biomedicine 92 (1): 135–143. doi:10.1016/j.cmpb.2008.06.010. 

External links

Wikimedia Commons has media related to Brightstat.

Wikiversity has learning materials about Brightstat

• Official website