# position of NaN values in terms of index df.loc[pandas.isna(df["b"]), :].index # position of NaN values in terms of rows that cotnain NaN df.loc[pandas.isna(df["b"]), :]