This java applet can be used to determine whether or not the means in two sample populations are significantly different.
JAVA applets must be supported.
Content Quality Concerns:
Definitions of discrete and continuous variables are not included in the documentation. All the examples of discrete variables are qualitative and those for continuous variables are quantitative. A quantitative example that is also discrete would help clarify the differences in the terms. Spearman Rank-Correlation was described in three places: in the flow chart as having factors of correlation rather than discrete or continuous, in the comparison charts as continuous and in the general description as either discrete or continuous.
Content Quality Strengths:
Documentation is concise and thorough. Organization of topics with links to terms and appropriate tests make the site easy to use. Included applets for the statistical tests make it easy for the student to analyze his/her own data. Sample data can be modified by the user, and calculations quickly redone, which allows the student to discover the effects of the modification.
Ease of Use Concerns:
The applets do not allow for the loading and saving of data sets, which limits their use to small data sets and training situations. Users that will need to use other tools to perform their tests will ultimately rely on those tools and abandon this site.
Ease of Use Strengths:
Use of links to definitions of terms is helpful to the user. Applets that include examples of each statistical method and allow the user to test his/her own data are excellent. A table of contents visible on the screen allows the user to effectively jump between topics.
Potential Effectiveness Concerns:
Feature(s) of Concern and Recommendations:It should be pointed out more clearly that the reader needs to obtain background knowledge on the tests somewhere else. Also (see ease of use), while the site supports the selection of an appropriate test, it may be cumbersome to execute the test.
Potential Effectiveness Strengths:
This is an excellent tool for the user who is designing an experiment and selecting appropriate statistical tests for analysis of the results. Thorough documentation makes it difficult to select an inappropriate test.
Potential Effectiveness Rating:
Source Code Available:
Source Code Available