Table of Contents


tzero - calculate Student’s t statistic versus zero mean (ERP statistics)


tzero [-o]opt [-h "Ascii header"]opt


tzero is used to calculate Student’s t statistic on a set of data for the null hypothesis that the population mean is 0. The data are taken from the standard input as ASCII numbers until an end of file is encountered, at which time the mean, standard deviation, standard error, number of data elements, t value (with degrees of freedom), and probability of that t arising under the null hypothesis are printed on the standard output. The input data can be arranged in relatively free form with one or more numbers per line and any number of lines. An end of file terminates the input. Usually the input data are redirected from a file, but can be typed in interactively if desired.

Normally tzero calculates the probability as a two-tailed test; the -o option causes tzero to calculate the probability as a one-tailed test (one half the two-tailed value).

The -h option enables the user to prepend an ASCII description to the output of tzero by immediately following the -h with a string surrounded by double quotes (as shown in the synopsis above), facilitating bookkeeping of multiple tests.


Jonathan C. Hansen UCSD ERPL


tzero complains if there is only one data point, as a test is impossible in this case.


The probability value is printed only to four decimal places; hence p values of .0000 can be emitted and should be interpreted as less than .0001.

Table of Contents