vis_anova_assumptions

vector containing dependent variable, datatype numeric

samples

vector containing independent variable, datatype factor

fact

conf.level

name of sample used in graphical output, dataype character , ""=default

samplename

name of sample used in graphical output, dataype character, ""=default

factorname

number indicating the amount by which plotting text and symbols should be scaled relative to the default. 1=default, 1.5 is 50% larger, 0.5 is 50% smaller, etc.

<code>vis_anova_assumptions</code> checks for normality of the standardised residuals of the anova both graphically by qq-plots as well as performing
 the Shapiro-Wilk-test <code>shapiro.test</code> and the Anderson-Darling-Test <code>ad.test</code>.
 <code>aov</code> further tests the homoscedacity of each factor level in <code>fact</code> with the <code>bartlett.test</code>.

Visualization of the most powerful statistical hypothesis test.
The function vistat() visualizes the statistical hypothesis testing between the
dependent variable (response) varsample and the independent variable (feature) varfactor.
The statistical hypothesis test (including the eventual corresponding post-hoc analysis) with
the highest statistical power fulfilling the assumptions of the corresponding test is chosen based on a decision tree.
A graph displaying the raw data accordingly to the chosen test is generated, the test statistics including eventual
post-hoc-analysis are returned. The automated workflow is especially suited for browser based interfaces to
server-based deployments of R. Implemented tests: lm(), t.test(), wilcox.test(), aov(), kruskal.test(), fisher.test(), chisqu.test().
Implemented tests to check the normal distribution of standardized residuals: shapiro.test() and ad.test().
Implemented post-hoc tests: TukeyHSD() for aov() and pairwise.wilcox.test() for kruskal.test().
For the comparison of averages, the following algorithm is implemented:
If the p-values of the standardized residuals of both shapiro.test() or ad.test() are smaller
than 1-conf.level, kruskal.test() resp. wilcox.test() are performed, otherwise the oneway.test()
and aov() resp. t.test() are performed and displayed. Exception:
If the sample size is bigger than 100, t.test() is always performed and wilcox.test() is never executed
(Lumley et al. (2002) <doi:10.1146/annurev.publhealth.23.100901.140546>).
For the test of independence of count data, Cochran's rule (Cochran (1954) <doi:10.2307/3001666>) is implemented:
If more than 20 percent of all cells have a count smaller than 5, fisher.test() is performed and displayed,
otherwise chisqu.test(). In both cases case an additional mosaic plot is generated.

Sabine Schilling

visStatistics

Automated Visualization of Statistical Tests

Peter Kauf

vis_anova_assumptions function

<dl><dt>samples</dt>
<dd>vector containing dependent variable, datatype numeric</dd>
<dt>fact</dt>
<dd>vector containing independent variable, datatype factor</dd>
<dt>conf.level</dt>
<dd>confidence level, 0.95=default</dd>
<dt>samplename</dt>
<dd>name of sample used in graphical output, dataype character , ""=default</dd>
<dt>factorname</dt>
<dd>name of sample used in graphical output, dataype character, ""=default</dd>
<dt>cex</dt>
<dd>number indicating the amount by which plotting text and symbols should be scaled relative to the default. 1=default, 1.5 is 50% larger, 0.5 is 50% smaller, etc.</dd></dl>

Arguments

Testing ANOVA assumptions — vis_anova_assumptions

<dl>

<dt>samples</dt>
<dd>vector containing dependent variable, datatype numeric</dd>


<dt>fact</dt>
<dd>vector containing independent variable, datatype factor</dd>


<dt>conf.level</dt>
<dd>confidence level, 0.95=default</dd>


<dt>samplename</dt>
<dd>name of sample used in graphical output, dataype character , ""=default</dd>


<dt>factorname</dt>
<dd>name of sample used in graphical output, dataype character, ""=default</dd>


<dt>cex</dt>
<dd>number indicating the amount by which plotting text and symbols should be scaled relative to the default. 1=default, 1.5 is 50% larger, 0.5 is 50% smaller, etc.</dd>

</dl>

Data engineering and BI courses are free!

vis_anova_assumptions: Testing ANOVA assumptions

Description

Usage

Value

Arguments

Examples