% Generated by roxygen2: do not edit by hand % Please edit documentation in R/ttest_class.R \name{ttest} \alias{ttest} \title{t-test} \usage{ ttest( alpha = 0.05, mtc = "fdr", factor_names, paired = FALSE, paired_factor = character(0), equal_variance = FALSE, conf_level = 0.95, control_group = NULL, ... ) } \arguments{ \item{alpha}{(numeric) The p-value cutoff for determining significance. The default is \code{0.05}.\cr} \item{mtc}{(character) Multiple test correction method. Allowed values are limited to the following: \itemize{ \item{\code{"bonferroni"}: Bonferroni correction in which the p-values are multiplied by the number of comparisons.}\item{\code{"fdr"}: Benjamini and Hochberg False Discovery Rate correction.}\item{\code{"none"}: No correction.}} The default is \code{"fdr"}.} \item{factor_names}{(character) The name of sample meta column(s) to use.} \item{paired}{(logical) Apply a paired t-test. The default is \code{FALSE}.\cr} \item{paired_factor}{(character) The factor name that encodes the sample id for pairing. The default is \code{character(0)}.} \item{equal_variance}{(logical) Equal variance. Allowed values are limited to the following: \itemize{ \item{\code{"TRUE"}: The variance of each group is treated as being equal using the pooled variance to estimate the variance.}\item{\code{"FALSE"}: The variance of each group is not assumed to be equal and the Welch (or Satterthwaite) approximation is used.}} The default is \code{FALSE}.\cr} \item{conf_level}{(numeric) The confidence level of the interval. The default is \code{0.95}.\cr} \item{control_group}{(character, NULL) The level name of the group used as the second group (where possible) when computing t-statistics. This ensures a positive t-statistic corresponds to an increase when compared to the control group. The default is \code{NULL}.} \item{...}{Additional slots and values passed to \code{struct_class}.} } \value{ A \code{ttest} object with the following \code{output} slots: \tabular{ll}{ \code{t_statistic} \tab (data.frame) The value of the calculate statistics which is converted to a p-value when compared to a t-distribution. \cr \code{p_value} \tab (data.frame) The probability of observing the calculated t-statistic. \cr \code{dof} \tab (numeric) The number of degrees of freedom used to calculate the test statistic. \cr \code{significant} \tab (data.frame) TRUE if the calculated p-value is less than the supplied threhold (alpha). \cr \code{conf_int} \tab (data.frame) Confidence interval for t statistic. \cr \code{estimates} \tab (data.frame) The group means estimated when computing the t-statistic. \cr } } \description{ A t-test compares the means of two factor levels. Multiple-test corrected p-values are used to indicate the significance of the computed difference for all features. } \section{Inheritance}{ A \code{ttest} object inherits the following \code{struct} classes: \cr\cr \verb{[ttest]} >> \verb{[model]} >> \verb{[struct_class]} } \examples{ M = ttest( alpha = 0.05, mtc = "fdr", factor_names = "V1", paired = FALSE, paired_factor = "NA", equal_variance = FALSE, conf_level = 0.95, control_group = NULL) M = ttest(factor_name='Class') }