diff --git a/pymc3/distributions/discrete.py b/pymc3/distributions/discrete.py
index c72af9b34a..e3462e6b8c 100644
--- a/pymc3/distributions/discrete.py
+++ b/pymc3/distributions/discrete.py
@@ -48,7 +48,7 @@ def __init__(self, n, p, *args, **kwargs):
         self.mode = tt.cast(tround(n * p), self.dtype)
 
     def random(self, point=None, size=None, repeat=None):
-        n, p = draw_values([self.n, self.p], point=point)
+        n, p = draw_values([self.n, self.p], point=point, size=size)
         return generate_samples(stats.binom.rvs, n=n, p=p,
                                 dist_shape=self.shape,
                                 size=size)
diff --git a/pymc3/distributions/distribution.py b/pymc3/distributions/distribution.py
index c0627254fb..0e7ba4f3c9 100644
--- a/pymc3/distributions/distribution.py
+++ b/pymc3/distributions/distribution.py
@@ -186,7 +186,7 @@ def __init__(self, logp, shape=(), dtype=None, testval=0, *args, **kwargs):
         self.logp = logp
 
 
-def draw_values(params, point=None):
+def draw_values(params, point=None, size=None):
     """
     Draw (fix) parameter values. Handles a number of cases:
 
@@ -215,10 +215,10 @@ def draw_values(params, point=None):
             for name, node in named_nodes.items():
                 if not isinstance(node, (tt.sharedvar.SharedVariable,
                                          tt.TensorConstant)):
-                    givens[name] = (node, _draw_value(node, point=point))
+                    givens[name] = (node, _draw_value(node, point=point, size=size))
     values = []
     for param in params:
-        values.append(_draw_value(param, point=point, givens=givens.values()))
+        values.append(_draw_value(param, point=point, givens=givens.values(), size=size))
     return values
 
 
@@ -247,7 +247,7 @@ def _compile_theano_function(param, vars, givens=None):
                     allow_input_downcast=True)
 
 
-def _draw_value(param, point=None, givens=None):
+def _draw_value(param, point=None, givens=None, size=None):
     """Draw a random value from a distribution or return a constant.
 
     Parameters
@@ -263,6 +263,8 @@ def _draw_value(param, point=None, givens=None):
     givens : dict, optional
         A dictionary from theano variables to their values. These values
         are used to evaluate `param` if it is a theano variable.
+    size: int, optional
+        Number of sampling
     """
     if isinstance(param, numbers.Number):
         return param
@@ -276,7 +278,7 @@ def _draw_value(param, point=None, givens=None):
         if point and hasattr(param, 'model') and param.name in point:
             return point[param.name]
         elif hasattr(param, 'random') and param.random is not None:
-            return param.random(point=point, size=None)
+            return param.random(point=point, size=size).mean(axis=0)
         else:
             if givens:
                 variables, values = list(zip(*givens))