diff --git a/docs/changelog.rst b/docs/changelog.rst
index b93a03d6..9b27e865 100644
--- a/docs/changelog.rst
+++ b/docs/changelog.rst
@@ -3,6 +3,15 @@
 What's new
 ##########
 
+v0.3.11 (unreleased)
+--------------------
+
+**Bugfixes**
+
+a. Passing a wrong ``tail`` argument to :py:func:`pingouin.correlation.corr` now *always* raises an error (see `PR 160 <https://github.com/raphaelvallat/pingouin/pull/160>`_).
+   In previous versions of ``pingouin``, using any ``method`` other than ``"pearson"`` and a wrong ``tail`` argument such as ``"two-tailed"`` or ``"both"``
+   (instead of the correct ``"two-sided"``) may have resulted in silently returning a one-sided p-value.
+
 v0.3.10 (February 2021)
 -----------------------
 
diff --git a/pingouin/correlation.py b/pingouin/correlation.py
index 12a13e15..3e16cc3d 100644
--- a/pingouin/correlation.py
+++ b/pingouin/correlation.py
@@ -508,6 +508,8 @@ def corr(x, y, tail='two-sided', method='pearson', **kwargs):
     y = np.asarray(y)
     assert x.ndim == y.ndim == 1, 'x and y must be 1D array.'
     assert x.size == y.size, 'x and y must have the same length.'
+    _msg = 'tail must be "two-sided" or "one-sided".'
+    assert tail in ['two-sided', 'one-sided'], _msg
 
     # Remove rows with missing values
     x, y = remove_na(x, y, paired=True)
@@ -529,7 +531,7 @@ def corr(x, y, tail='two-sided', method='pearson', **kwargs):
     elif method == 'skipped':
         r, pval, outliers = skipped(x, y, **kwargs)
     else:
-        raise ValueError('Method not recognized.')
+        raise ValueError(f'Method "{method}" not recognized.')
 
     if np.isnan(r):
         # Correlation failed -- new in version v0.3.4, instead of raising an