normal_test

Conducts normality tests for a specified variable, optionally by group. Supports automatic testing and interactive visualization.

Streamlines the analysis of clinical data by automatically selecting appropriate statistical descriptions and inference methods based on variable types. For method details see Motulsky H J (2016) <https://www.graphpad.com/guides/prism/10/statistics/index.htm> and d'Agostino R B (1971) <doi:10.1093/biomet/58.2.341>.

Xiang Li

tidysummary

An Elegant Approach to Summarizing Clinical Data

normal_test function

<dl><dt>data</dt>
<dd>A data frame containing the variables to be tested.</dd>
<dt>var</dt>
<dd>A character string specifying the numeric variable in <code>data</code> to test.</dd>
<dt>group</dt>
<dd>A character string specifying the grouping variable in <code>data</code>. If <code>NULL</code>, treated as one group.</dd>
<dt>norm</dt>
<dd>Control parameter for test behavior. Accepts:<ul>
<li><code>'auto'</code>: Automatically decide based on p-values, but the same as <code>'ask'</code> when n &gt; 1000, default</li>
<li><code>'ask'</code>: Show p-values, plots QQ plots and prompts for decision</li>
<li><code>TRUE</code>/<code>'true'</code>: Always returns <code>TRUE</code></li>
<li><code>FALSE</code>/<code>'false'</code>: Always returns <code>FALSE</code></li>
</ul></dd></dl>

Arguments

Automatically selects test based on sample size per group:<ul>
<li>n &lt; 3: Too small, assuming non-normal</li>
<li>(3, 50] Shapiro-Wilk test</li>
<li>(50, 1000]: D'Agostino Chi2 test, instead of Kolmogorov-Smirnov test</li>
<li>n &gt; 1000: Show p-values, plots QQ plots and prompts for decision</li>
</ul>

Methodology for p-values

Perform normality test on a variable — normal_test

<dl>

<dt>data</dt>
<dd>A data frame containing the variables to be tested.</dd>


<dt>var</dt>
<dd>A character string specifying the numeric variable in <code>data</code> to test.</dd>


<dt>group</dt>
<dd>A character string specifying the grouping variable in <code>data</code>. If <code>NULL</code>, treated as one group.</dd>


<dt>norm</dt>
<dd>Control parameter for test behavior. Accepts:<ul>
<li><code>'auto'</code>: Automatically decide based on p-values, but the same as <code>'ask'</code> when n &gt; 1000, default</li>
<li><code>'ask'</code>: Show p-values, plots QQ plots and prompts for decision</li>
<li><code>TRUE</code>/<code>'true'</code>: Always returns <code>TRUE</code></li>
<li><code>FALSE</code>/<code>'false'</code>: Always returns <code>FALSE</code></li>
</ul></dd>

</dl>

Automatically selects test based on sample size per group:<ul>
<li>n &lt; 3: Too small, assuming non-normal</li>
<li>(3, 50] Shapiro-Wilk test</li>
<li>(50, 1000]: D'Agostino Chi2 test, instead of Kolmogorov-Smirnov test</li>
<li>n &gt; 1000: Show p-values, plots QQ plots and prompts for decision</li>
</ul>

normal_test: Perform normality test on a variable

Description

Usage

Value

Arguments

Methodology for p-values

Examples