diff --git a/python/cugraph/cugraph/datasets/dataset.py b/python/cugraph/cugraph/datasets/dataset.py
index 9817d15dacb..15c30700fc3 100644
--- a/python/cugraph/cugraph/datasets/dataset.py
+++ b/python/cugraph/cugraph/datasets/dataset.py
@@ -174,7 +174,7 @@ def get_edgelist(self, download=False, reader="cudf"):
         reader : 'cudf' or 'pandas' (default='cudf')
             The library used to read a CSV and return an edgelist DataFrame.
         """
-        if self._edgelist is None:
+        if self._edgelist is None or not isinstance(self._edgelist, cudf.DataFrame):
             full_path = self.get_path()
             if not full_path.is_file():
                 if download:
@@ -223,7 +223,9 @@ def get_dask_edgelist(self, download=False):
             Automatically download the dataset from the 'url' location within
             the YAML file.
         """
-        if self._edgelist is None:
+        if self._edgelist is None or not isinstance(
+            self._edgelist, dask_cudf.DataFrame
+        ):
             full_path = self.get_path()
             if not full_path.is_file():
                 if download:
@@ -286,7 +288,7 @@ def get_graph(
             for certain algorithms, such as pagerank.
         """
         if self._edgelist is None:
-            self.get_edgelist(download)
+            self.get_edgelist(download=download)
 
         if create_using is None:
             G = Graph()
@@ -351,7 +353,7 @@ def get_dask_graph(
             for certain algorithms.
         """
         if self._edgelist is None:
-            self.get_dask_edgelist(download)
+            self.get_dask_edgelist(download=download)
 
         if create_using is None:
             G = Graph()
@@ -367,7 +369,7 @@ def get_dask_graph(
                 f"{type(create_using)}"
             )
 
-        if len(self.metadata["col_names"]) > 2 and not (ignore_weights):
+        if len(self.metadata["col_names"]) > 2 and not ignore_weights:
             G.from_dask_cudf_edgelist(
                 self._edgelist,
                 source=self.metadata["col_names"][0],
diff --git a/python/cugraph/cugraph/tests/utils/test_dataset.py b/python/cugraph/cugraph/tests/utils/test_dataset.py
index 9331cefcfda..fae89e02002 100644
--- a/python/cugraph/cugraph/tests/utils/test_dataset.py
+++ b/python/cugraph/cugraph/tests/utils/test_dataset.py
@@ -20,7 +20,11 @@
 import pytest
 
 import cudf
+import cugraph
 import dask_cudf
+from cugraph import datasets
+from cugraph.dask.common.mg_utils import is_single_gpu
+from cugraph.datasets import karate
 from cugraph.structure import Graph
 from cugraph.testing import (
     RAPIDS_DATASET_ROOT_DIR_PATH,
@@ -29,8 +33,7 @@
     SMALL_DATASETS,
     BENCHMARKING_DATASETS,
 )
-from cugraph import datasets
-from cugraph.dask.common.mg_utils import is_single_gpu
+
 
 # Add the sg marker to all tests in this module.
 pytestmark = pytest.mark.sg
@@ -38,6 +41,7 @@
 
 ###############################################################################
 # Fixtures
+###############################################################################
 
 
 # module fixture - called once for this module
@@ -201,7 +205,9 @@ def test_reader_dask(dask_client, dataset):
 @pytest.mark.parametrize("dataset", ALL_DATASETS)
 def test_get_edgelist(dataset):
     E = dataset.get_edgelist(download=True)
+
     assert E is not None
+    assert isinstance(E, cudf.DataFrame)
 
 
 @pytest.mark.skipif(is_single_gpu(), reason="skipping MG testing on Single GPU system")
@@ -209,13 +215,16 @@ def test_get_edgelist(dataset):
 @pytest.mark.parametrize("dataset", ALL_DATASETS)
 def test_get_dask_edgelist(dask_client, dataset):
     E = dataset.get_dask_edgelist(download=True)
+
     assert E is not None
+    assert isinstance(E, dask_cudf.DataFrame)
 
 
 @pytest.mark.parametrize("dataset", ALL_DATASETS)
 def test_get_graph(dataset):
     G = dataset.get_graph(download=True)
     assert G is not None
+    assert isinstance(G, cugraph.Graph)
 
 
 @pytest.mark.skipif(is_single_gpu(), reason="skipping MG testing on Single GPU system")
@@ -224,12 +233,14 @@ def test_get_graph(dataset):
 def test_get_dask_graph(dask_client, dataset):
     G = dataset.get_dask_graph(download=True)
     assert G is not None
+    # TODO Check G is a DistributedGraph
 
 
 @pytest.mark.parametrize("dataset", ALL_DATASETS)
 def test_metadata(dataset):
     M = dataset.metadata
     assert M is not None
+    assert isinstance(M, dict)
 
 
 @pytest.mark.parametrize("dataset", ALL_DATASETS)
@@ -346,32 +357,19 @@ def test_ctor_with_datafile():
     assert ds.get_path() == karate_csv
 
 
-def test_unload():
-    email_csv = RAPIDS_DATASET_ROOT_DIR_PATH / "email-Eu-core.csv"
+@pytest.mark.parametrize("dataset", [karate])
+def test_unload(dataset):
+    assert dataset._edgelist is None
 
-    ds = datasets.Dataset(
-        csv_file=email_csv.as_posix(),
-        csv_col_names=["src", "dst", "wgt"],
-        csv_col_dtypes=["int32", "int32", "float32"],
-    )
+    dataset.get_edgelist()
+    assert dataset._edgelist is not None
+    dataset.unload()
+    assert dataset._edgelist is None
 
-    # FIXME: another (better?) test would be to check free memory and assert
-    # the memory use increases after get_*(), then returns to the pre-get_*()
-    # level after unload(). However, that type of test may fail for several
-    # reasons (the device being monitored is accidentally also being used by
-    # another process, and the use of memory pools to name two). Instead, just
-    # test that the internal members get cleared on unload().
-    assert ds._edgelist is None
-
-    ds.get_edgelist()
-    assert ds._edgelist is not None
-    ds.unload()
-    assert ds._edgelist is None
-
-    ds.get_graph()
-    assert ds._edgelist is not None
-    ds.unload()
-    assert ds._edgelist is None
+    dataset.get_graph()
+    assert dataset._edgelist is not None
+    dataset.unload()
+    assert dataset._edgelist is None
 
 
 @pytest.mark.parametrize("dataset", ALL_DATASETS)