eelbrain.testnd.ttest_rel

class eelbrain.testnd.ttest_rel(y: Union[eelbrain._data_obj.NDVar, str], x: Union[eelbrain._data_obj.Factor, eelbrain._data_obj.Interaction, eelbrain._data_obj.NestedEffect, str, eelbrain._data_obj.NDVar], c1: Union[str, Tuple[str, ...]] = None, c0: Union[str, Tuple[str, ...]] = None, match: Union[eelbrain._data_obj.Factor, eelbrain._data_obj.Interaction, eelbrain._data_obj.NestedEffect, str] = None, sub: Union[eelbrain._data_obj.Var, numpy.ndarray, str] = None, ds: eelbrain._data_obj.Dataset = None, tail: int = 0, samples: int = 10000, pmin: float = None, tmin: float = None, tfce: Union[float, bool] = False, tstart: float = None, tstop: float = None, parc: str = None, force_permutation: bool = False, **criteria)

Mass-univariate related samples t-test

The test data can be specified in two forms:

• In long form, with y supplying the data, x specifying condition for each case and match determining which cases are related.
• In wide/repeated measures form, with y and x both supplying data with matching case order.

Parameters:

y : NDVar

Dependent variable.

x : categorial | NDVar

Model containing the cells which should be compared, or NDVar to which y should be compared. In the latter case, the next three parameters are ignored.

c1 : str | tuple | None

Test condition (cell of x). c1 and c0 can be omitted if x only contains two cells, in which case cells will be used in alphabetical order.

c0 : str | tuple | None

Control condition (cell of x).

match : categorial

Units within which measurements are related (e.g. ‘subject’ in a within-subject comparison).

sub : index

Perform the test with a subset of the data.

ds : None | Dataset

If a Dataset is specified, all data-objects can be specified as names of Dataset variables.

tail : 0 | 1 | -1

Which tail of the t-distribution to consider: 0: both (two-tailed, default); 1: upper tail (one-tailed); -1: lower tail (one-tailed).

samples : int

Number of samples for permutation test (default 10,000).

pmin : None | scalar (0 < pmin < 1)

Threshold for forming clusters: use a t-value equivalent to an uncorrected p-value.

tmin : scalar

Threshold for forming clusters as t-value.

tfce : bool | scalar

Use threshold-free cluster enhancement. Use a scalar to specify the step of TFCE levels (for tfce is True, 0.1 is used).

tstart : scalar

Start of the time window for the permutation test (default is the beginning of y).

tstop : scalar

Stop of the time window for the permutation test (default is the end of y).

parc : str

Collect permutation statistics for all regions of the parcellation of this dimension. For threshold-based test, the regions are disconnected.

force_permutation: bool

Conduct permutations regardless of whether there are any clusters.

mintime : scalar

Minimum duration for clusters (in seconds).

minsource : int

Minimum number of sources per cluster.

Notes

Also known as dependent t-test, paired t-test or repeated measures t-test. In the permutation cluster test, permutations are done within the categories of match. Cases with zero variance are set to t=0.

Attributes:

c1_mean : NDVar

Mean in the c1 condition.

c0_mean : NDVar

Mean in the c0 condition.

clusters : None | Dataset

For cluster-based tests, a table of all clusters. Otherwise a table of all significant regions (or None if permutations were omitted). See also the find_clusters() method.

difference : NDVar

Difference between the mean in condition c1 and condition c0.

p : NDVar | None

Map of p-values corrected for multiple comparison (or None if no correction was performed).

p_uncorrected : NDVar

Map of p-values uncorrected for multiple comparison.

t : NDVar

Map of t-values.

tfce_map : NDVar | None

Map of the test statistic processed with the threshold-free cluster enhancement algorithm (or None if no TFCE was performed).

n : int

Number of cases.

Methods

cluster(self, cluster_id)	Retrieve a specific cluster as NDVar
compute_probability_map(self, **sub)	Compute a probability map
find_clusters(self[, pmin, maps])	Find significant regions or clusters
find_peaks(self)	Find peaks in a threshold-free cluster distribution
info_list(self[, computation])	List with information about the test
masked_c1(self[, p])	c1 map masked by significance of the c1-c0 difference
masked_difference(self[, p, name])	Difference map masked by significance
masked_parameter_map(self[, pmin])	Create a copy of the parameter map masked by significance