Resource Library

Analysis of variance (ANOVA) is used to test hypotheses about differences between two or more means. The t-test based on the standard error of the difference between two means can only be used to test differences between two means. When there are more than two means, it is possible to compare each mean with each other mean using t-tests. However, conducting multiple t-tests can lead to severe inflation of the Type I error rate. (Click here to see why) Analysis of variance can be used to test differences among several means for significance without increasing the Type I error rate. This chapter covers designs with between-subject variables. 

When an experimenter is interested in the effects of two or more independent variables, it is usually more efficient to manipulate these variables in one experiment than to run a separate experiment for each variable. Moreover, only in experiments with more than one independent variable is it possible to test for interactions among variables.  Experimental designs in which every level of every variable is paired with every level of every other variable are called factorial designs. 

Within-subject designs are designs in which one or more of the independent variables are within-subject variables. Within-subjects designs are often called repeated-measures designs since within-subjects variables always involve taking repeated measurements from each subject. Within-subject designs are extremely common in psychological and biomedical research.

The JCCKit is a small library and flexible framework for creating scientific charts and plots works on java platform. 

When two variables are related, it is possible to predict a person's score on one variable from their score on the second variable with better than chance accuracy. This section describes how these predictions are made and what can be learned about the relationship between the variables by developing a prediction equation.

This resource gives a thorough definition of confidence intervals. It shows the user how to compute a confidence interval and how to interpret them. It goes into detail on how to construct a confidence interval for the difference between means, correlations, and proportions. It also gives a detailed explanation of Pearson's correlation. It also includes exercises for the user.

JFreeReport is a free Java report library. It has the following features: full on-screen print preview; data obtained via Swing's TableModel interface (making it easy to print data directly from your application); XML-based report definitions; output to the screen, printer or various export formats (PDF, HTML, CSV, Excel, plain text); support for servlets (uses the JFreeReport extensions) complete source code included (subject to the GNU Lesser General Public Licence); extensive source code documentation.

This chapter discusses a collection of tests called distribution-free tests, or nonparametric tests, that do not make any assumptions about the distribution from which the numbers were sampled. The main advantage of distribution-free tests is that they provide more power than traditional tests when the samples are from highly-skewed distributions. 

JFreeChart is a free Java class library for generating charts, including: pie charts (2D and 3D); bar charts (regular and stacked, with an optional 3D effect); line and area charts; scatter plots and bubble charts; time series, high/low/open/close charts and candle stick charts; combination charts; Pareto charts; Gantt charts; wind plots, meter charts and symbol charts; wafer map charts;

Measures of the size of an effect based on the degree of overlap between groups usually involve calculating the proportion of the variance that can be explained by differences between groups. This resource outlines different approaches to measuring this proportion.