diff --git a/src/sorcha/modules/PPStats.py b/src/sorcha/modules/PPStats.py
index 1307842e..324108dd 100644
--- a/src/sorcha/modules/PPStats.py
+++ b/src/sorcha/modules/PPStats.py
@@ -44,6 +44,9 @@ def stats(observations, statsfilename, outpath, configs):
         linked = group_by["object_linked"].agg("all").to_frame("object_linked")
         date_linked = group_by["date_linked_MJD"].agg("first").to_frame("date_linked_MJD")
         joined_stats = num_obs.join([mag, phase_deg, linked, date_linked])
+    elif configs["SSP_linking_on"]:
+        date_linked = group_by["date_linked_MJD"].agg("first").to_frame("date_linked_MJD")
+        joined_stats = num_obs.join([mag, phase_deg, date_linked])
     else:
         joined_stats = num_obs.join([mag, phase_deg])
 
diff --git a/tests/sorcha/test_PPStats.py b/tests/sorcha/test_PPStats.py
index eebbec32..f60ab6e4 100644
--- a/tests/sorcha/test_PPStats.py
+++ b/tests/sorcha/test_PPStats.py
@@ -123,3 +123,54 @@ def test_PPStats_nolinking(tmp_path):
     # the previous test checks all rows so it's fine to just check one here, this test is mostly to make
     # sure that the stats file works correctly if linking is off
     assert_equal(expected_row_one, stats_df.iloc[0].values)
+
+
+def test_PPStats_justlinking(tmp_path):
+    # tests behaviour when linking is on but drop_unlinked=True
+
+    ObjID = (["object_one"] * 10) + (["object_two"] * 5)
+    optFilter = (["r"] * 6) + (["g"] * 4) + (["r"] * 5)
+    trailedSourceMag = np.concatenate(
+        (np.linspace(18, 21, 6), np.linspace(19, 22, 4), np.linspace(20, 23, 5))
+    )
+    phase_deg = np.concatenate((np.linspace(3, 10, 6), np.linspace(4, 11, 4), np.linspace(5, 10, 5)))
+    obj_date = np.array(([666.0] * 10) + ([np.nan] * 5), dtype=object)
+
+    test_dict = {
+        "ObjID": ObjID,
+        "optFilter": optFilter,
+        "trailedSourceMag": trailedSourceMag,
+        "phase_deg": phase_deg,
+        "date_linked_MJD": obj_date,
+    }
+    test_df = pd.DataFrame(test_dict)
+
+    configs = {"SSP_linking_on": True, "drop_unlinked": True}
+
+    filename_stats = "test_stats"
+    stats(test_df, filename_stats, tmp_path, configs)
+
+    stats_df = pd.read_csv(os.path.join(tmp_path, filename_stats + ".csv"))
+
+    assert len(stats_df) == 3
+
+    expected_columns = np.array(
+        [
+            "ObjID",
+            "optFilter",
+            "number_obs",
+            "min_apparent_mag",
+            "max_apparent_mag",
+            "median_apparent_mag",
+            "min_phase",
+            "max_phase",
+            "date_linked_MJD",
+        ],
+        dtype=object,
+    )
+
+    assert_equal(expected_columns, stats_df.columns.values)
+
+    expected_row_one = np.array(["object_one", "g", 4, 19.0, 22.0, 20.5, 4.0, 11.0, 666.0], dtype=object)
+
+    assert_equal(expected_row_one, stats_df.iloc[0].values)