Single Degree of Freedom Contrasts:  When using factorial designs with a factor that has more than two levels, it is rare for investigators to examine overall F ratios for the effects and stop there.  Rather, follow-up tests are conducted to address more focused questions.  For main effects, this often takes the form of a Tukey procedure or a Bonferroni based procedure.

Using the SPSS GLM Program:  The major method that SPSS uses for testing interaction contrasts is the LMATRIX subcommand in its GLM program.  You determine the contrast coefficients you need to use, write the appropriate SPSS syntax and then add the syntax to an SPSS GLM program.  The process is cumbersome and time consuming.  ZumaStat makes it painless.  Using a clever but simple input strategy, you indicate to ZumaStat what contrast you want.  It can be a main effect contrast, a simple main effect contrast or an interaction contrast.  ZumaStat generates the coefficients you need to use and then provides you with the syntax that you can cut and paste directly into SPSS.  ZumaStat will do this for two factor, three factor and four factor designs, with or without covariates.  This routine does not allow complex contrast coding (such as orthogonal contrasts) but rather focuses on single degree of freedom contrasts that are typically of interest to researchers.

Equivalence Testing:  It is well known that in null hypothesis testing, one can never accept the null hypothesis.  This means that you can never state that two or more groups are equivalent on some outcome.  Yet investigators often desire to assert equivalence. There is a large literature on statistical equivalence testing that addresses this issue. 

Meta-Analysis:  ZumaStat provides for meta analyses of raw mean differences as well as the effect size measure for mean difference, d.  It applies both fixed effects and random effects models for averaging these statistics as well as tests for homogeneity of effect size. I n addition, it provides programs for conducting contrast analyses of different study groupings to determine if the average effect size for one group of studies differs from the average effect size for another group of studies.  ZumaStat also provides utilities for weighted least squares regression models predicting effect sizes from study characteristics that are continuous or categorical in nature.  Finally, ZumaStat offers an extensive set of programs for conducting power analysis in meta-analysis (see the complete list of programs in the 'List of Programs' section). These include power analysis for the test of average d statistics, power analysis for contrasts between groups of studies on average d statistics, power analysis of WLS regression analyses of d statistics and power analysis for the Q test of homogeneity for d statistics for a fixed effects model.

Power Analysis: ZumaStat offers power analysis for several common statistical tests. Importantly, it permits you to conduct a power analysis on specific contrasts within complex factorial designs. Many researchers conduct power analysis on the omnibus effects for a factor. But interest usually is focused on what happens at the level of the pairwise contrasts within that factor. ZumaStat helps to ensure that you will have adequate statistical power for such tests. The utilities allow you to either specify a desired level of power and an effect size and determine the sample size you will need, or you can specify a sample sizer and an effect size and obtain the statistical power associated with it. ZumaStat offers power analysis for one sample t test, correlated groups t test, one way ANOVA, two factor ANOVA, three factor ANOVA, one way analysis of covariance, two factor analysis of covariance, three factor analysis of covariance, between-within designs, cluster randomized designs, multi-site cluster designs, and designs with growth curves.

Analysis of Precision: Some researchers focus not on hypothesis testing but rather on magnitude estimation. The focus in this approach is estimating the magnitude of a difference rather than whether that difference is zero or not. When designing a study, you want to make sure that your magnitude estimates will be sufficiently precise and not subject to too much random error. ZumaStat offers utilities for determining sample sizes you should use to minimize sampling error. You provide a confidence interval width that you want to achieve and ZumaStat suggests a sample size for you.

General Mean Contrasts:  ZumaStat offers a general routine for conducting single degree of freedom contrasts for means from multiple groups.  You enter the contrast coefficients and the means and ZumaStat does the rest.  ZumaStat provides both significance tests and confidence intervals for the case of homogenous variances and heterogeneous variances.  You can analyze up to 1,000 different groups. 

Analysis of Trimmed Means.  ZumaStat provides utilities to calculate trimmed means and Winsorized standard deviations in an SPSS data file.  It also has a utility for an independent groups t test of trimmed means, a dependent groups t test of trimmed means, a one sample t test and linear contrasts of trimmed means.

Repeated Measure Designs:  ZumaStat describes shortcuts you can use to effectively analyze data in repeated measure factorial designs, taking advantage of other ZumaStat subroutines.  These are described in the extensive Help menus of ZumaStat. 

Regions of Significance: This program performs a Johnson-Neyman analysis for interactions between a categorical variable and a continuous variable.  Suppose you are comparing the means of two groups (e.g., experimental versus control) on an outcome variable, Y.  There is a continuous variable, Z, that interacts with group membership in the sense that the mean difference between the two groups varies depending on the value of Z (e.g., the mean difference might vary as a function of social class).  The program calculates an estimate of the value of Z where the population means for the two groups are equal. It also forms a confidence interval about this value, to define a region of significance.  When Z is larger than the upper limit, one can be confident that the population means are not equal in one direction and when Z is less than the lower limit, one can be confident that the population means are not equal in the other direction. 

Regression to the Mean: ZumaStat offers utility that allows you to estimate how much regression to the mean is operating for an experimental versus control group study in which there are pre-existing group differences on the outcome variable and the analysis is based on examining posttest mean differences after statistically adjusting for the pre-test differences.

Confidence Interval for a Standard Deviation:  ZumaStat calculates the confidence interval for a standard deviation and also a confidence interval for a variance ratio.  The latter is useful for evaluating homogeneity of variance assumptions.

Multiple Comparison Methods: ZumaStat offers a wide range of multiple comparison procedures as applied to summary statistics.  These include Tukey HSD tests, Tukey LSD tests, Tukey-Kramer tests, REGW Q tests, the Games and Howell test, Fisher-Hayter tests and Sidak tests.  In addition, ZumaStat provides the Holm modified Bonferroni method, the Hochberg modified Bonferroni method, the False Discovery Rate approach and the Scheffe test.

Confidence Interval for a Mean and One Sample t Test:  ZumaStat calculates the confidence interval for a mean using the sample size and standard deviation as input.  It also performs a significance test against a hypothesized population value.

Confidence Interval for a Mean Difference and t Tests:  ZumaStat calculates both a significance test and a confidence interval for testing the difference between two means.  Both the case of independent and dependent means are covered.  For the former, input is the means, sample sizes, and standard deviations of the two groups.  For the latter, a correlation coefficient is also required.  Indices of effect size are also reported.

One Way Analysis of Variance:  ZumaStat calculates a one-way, between-subjects analysis of variance from group means, sample sizes, and standard deviations.  Effect size indices are also reported. 

Pooled Mean:  ZumaStat calculates a pooled mean from several different group means, both weighted and unweighted by the group sample sizes.

Common Language Effect Size:  ZumaStat calculates the Common Language Effect Size (CLES) for characterizing the magnitude of a mean difference for two groups.  This is an intuitive and increasingly popular index of effect size.  Both the cases of independent groups and dependent groups are available.  The index tells you the probability that a randomly selected individual from one group will have a higher score than a randomly selected individual from another group.  A nonparametric version of the index, called the probability of superiority, also is included in ZumaStat.

Confidence Intervals for Standardized Effect Sizes:  ZumaStat calculates confidence intervals for percent of variance accounted for statistics in analysis of variance and t test situations as well as for the d statistic.

Excel Graphs:  ZumaStat will create Excel graphs for one way analysis of variance as well as two way plots for factorial analysis of variance based on the means that you provide.  The creation of the charts is quick and painless and ready to be cut and pasted into Word or some other word processor.  All you need is Excel.

Summary Table from Traditional Statistics: Most journal articles reports means, sample sizes and F ratios, but do not provide you with the more informative summary table for ANOVA. ZumaStat will recreate a summary table from the information traditionally reported in a journal report.