diff --git a/measurements/text_duplicates/text_duplicates.py b/measurements/text_duplicates/text_duplicates.py
index 013646440..a1f0a85d8 100644
--- a/measurements/text_duplicates/text_duplicates.py
+++ b/measurements/text_duplicates/text_duplicates.py
@@ -82,7 +82,7 @@ def _compute(self, data, list_duplicates=False):
         if list_duplicates == True:
             logger.warning("This functionality can be memory-intensive for large datasets!")
             n_dedup = len(set([get_hash(d) for d in data]))
-            c = Counter(data)
+            c = Counter(d.strip() for d in data)
             duplicates = {k: v for k, v in c.items() if v > 1}
             return {"duplicate_fraction": 1 - (n_dedup / len(data)), "duplicates_dict": duplicates}
         else:
diff --git a/tests/test_metric_common.py b/tests/test_metric_common.py
index 014dc0b32..b6d711ec8 100644
--- a/tests/test_metric_common.py
+++ b/tests/test_metric_common.py
@@ -225,3 +225,11 @@ def test_seqeval_raises_when_incorrect_scheme():
     error_message = f"Scheme should be one of [IOB1, IOB2, IOE1, IOE2, IOBES, BILOU], got {wrong_scheme}"
     with pytest.raises(ValueError, match=re.escape(error_message)):
         metric.compute(predictions=[], references=[], scheme=wrong_scheme)
+
+
+def test_text_duplicates_lists_the_duplicates_it_counts():
+    measurement = load(os.path.join("measurements", "text_duplicates"))
+    # the fraction ignores surrounding whitespace, so the listed duplicates have to as well
+    results = measurement.compute(data=["hello sun", "hello sun ", "hello moon"], list_duplicates=True)
+    assert results["duplicate_fraction"] == pytest.approx(1 / 3)
+    assert results["duplicates_dict"] == {"hello sun": 2}