diff --git a/notebooks/q29_cleaning.ipynb b/notebooks/q29_cleaning.ipynb
index fb484d9..cb6ce57 100644
--- a/notebooks/q29_cleaning.ipynb
+++ b/notebooks/q29_cleaning.ipynb
@@ -23,7 +23,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "0b6dc5f5-e2df-47e2-ab43-3c49600290b1",
    "metadata": {},
    "outputs": [],
@@ -47,7 +47,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "250889a9-f63c-45c1-9bf6-eb0bd1073a89",
    "metadata": {},
    "outputs": [],
@@ -87,7 +87,9 @@
    "cell_type": "code",
    "execution_count": null,
    "id": "27fe45d8-cfbd-47e2-b295-939cc8f4372e",
-   "metadata": {},
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [],
    "source": [
     "# Full Model Pipeline\n",
@@ -120,8 +122,10 @@
     "print('Filtering to new IDs...', end = '')\n",
     "# grab only the IDs of current interest\n",
     "completed_ids = df_done.id.unique()\n",
-    "#df = df_open[~df_open.id.isin(completed_ids)].reset_index(drop=True)\n",
-    "df = df_open.copy()\n",
+    "df = df_open[~df_open.id.isin(completed_ids)].reset_index(drop=True)\n",
+    "\n",
+    "# remove empty responses\n",
+    "df = df[~ (pd.isnull(df.aq29lang) | (df.aq29lang.str.len()==0))].reset_index(drop=True)\n",
     "\n",
     "df['response'] = df['aq29lang']\n",
     "print('Done.')\n",