I have a dataset with about 5688 records of subjects that have needed sedations for a procedure. The records have information such as the procedure type, the type of physician/professional administering the sedation, etc...So some doctors have more records because they do more procedures. I want to simulate (or re-create) this dataset, so that there are 146164 records or as close to this number of records as possible. I want to maintain the 'weighting' of records entered by the sedation doctor. Here's an example data. Let me describe what I desire by using this small sample. So say I had these 10 records, I want to have an output dataset of 146 records, resampled from this dataset of 10. I want to keep the 'weighting' or proportion of the physician type where the Anesthesiologist accounts for a majority of the records, in this example 7 out of 10 (70%). Procedure Sedative route Physician type Laceration/suture IV Nurse Anesthesiologist Dental procedure IV Dental surgeon Dental surgery IV Dental surgeion MRI IV Anesthesiologist Gastro endoscopy - U IV Anesthesiologist Gastro endoscopy – L IV Anesthesiologist Other IV Anesthesiologist Lumbar puncture IV Anesthesiologist MRI IV Anesthesiologist CT scan IV Anesthesiologist
... View more