diff --git a/dataprofiler/profilers/numerical_column_stats.py b/dataprofiler/profilers/numerical_column_stats.py
index 6086b575..fa0666a6 100644
--- a/dataprofiler/profilers/numerical_column_stats.py
+++ b/dataprofiler/profilers/numerical_column_stats.py
@@ -13,7 +13,7 @@
 import pandas as pd
 import scipy.stats
 
-from . import histogram_utils, profiler_utils
+from . import float_column_profile, histogram_utils, profiler_utils
 from .base_column_profilers import BaseColumnProfiler
 from .profiler_options import NumericalOptions
 
@@ -710,6 +710,9 @@ def _preprocess_for_calculate_psi(
                     entity_count_per_bin=self_histogram["bin_counts"],
                     bin_edges=self_histogram["bin_edges"],
                     suggested_bin_count=num_psi_bins,
+                    is_float_histogram=isinstance(
+                        self, float_column_profile.FloatColumn
+                    ),
                     options={
                         "min_edge": min_min_edge,
                         "max_edge": max_max_edge,
@@ -731,6 +734,9 @@ def _preprocess_for_calculate_psi(
                     entity_count_per_bin=other_histogram["bin_counts"],
                     bin_edges=other_histogram["bin_edges"],
                     suggested_bin_count=num_psi_bins,
+                    is_float_histogram=isinstance(
+                        self, float_column_profile.FloatColumn
+                    ),
                     options={
                         "min_edge": min_min_edge,
                         "max_edge": max_max_edge,
@@ -1360,7 +1366,12 @@ def _update_histogram(self, df_series: pd.Series) -> None:
         self._stored_histogram["total_loss"] += histogram_loss
 
     def _regenerate_histogram(
-        self, entity_count_per_bin, bin_edges, suggested_bin_count, options=None
+        self,
+        entity_count_per_bin,
+        bin_edges,
+        suggested_bin_count,
+        is_float_histogram,
+        options=None,
     ) -> tuple[dict[str, np.ndarray], float]:
 
         # create proper binning
@@ -1372,6 +1383,11 @@ def _regenerate_histogram(
             new_bin_edges = np.linspace(
                 options["min_edge"], options["max_edge"], suggested_bin_count + 1
             )
+
+        # if it's not a float histogram, then assume it only contains integer values
+        if not is_float_histogram:
+            bin_edges = np.round(bin_edges)
+
         return self._assimilate_histogram(
             from_hist_entity_count_per_bin=entity_count_per_bin,
             from_hist_bin_edges=bin_edges,
@@ -1417,11 +1433,6 @@ def _assimilate_histogram(
 
             bin_edge = from_hist_bin_edges[bin_id : bin_id + 3]
 
-            # if we know not float, we can assume values in bins are integers.
-            is_float_profile = self.__class__.__name__ == "FloatColumn"
-            if not is_float_profile:
-                bin_edge = np.round(bin_edge)
-
             # loop until we have a new bin which contains the current bin.
             while (
                 bin_edge[0] >= dest_hist_bin_edges[new_bin_id + 1]
@@ -1513,6 +1524,7 @@ def _histogram_for_profile(
             entity_count_per_bin=bin_counts,
             bin_edges=bin_edges,
             suggested_bin_count=suggested_bin_count,
+            is_float_histogram=isinstance(self, float_column_profile.FloatColumn),
         )
 
     def _get_best_histogram_for_profile(self) -> dict: