from pyspark.sql.functions import lit df = sqlContext.createDataFrame( [(1, "a", 23.0), (3, "B", -23.0)], ("x1", "x2", "x3")) df_with_x4 = df.withColumn("x4", lit(0)) df_with_x4.show()