So in the base exam prep guide, the first data set combine method is the One-to-One Reading (from here on OTOR), where the second data set overrides the first data set in the same named column. Basic logic is shown below.     In the example, they used two data sets cert.patients and cert.measure.  And they used the IF condition age < 60. Anyways, the patients data set is not sorted, so you'll have to sort it by ID first to get the same result as the book. So the code should be proc sort data=spg.patients;by id; run; proc print data=spg.patients;run; proc print data=spg.measure;run; data one2one; set spg.patients; if age<60; set spg.measure; run; proc print data=one2one;run; And result will be..   Anyways regardless, where is the point of OTOR? The only reason why the combined result is correct (correct as in that all 9 obs in the combined set have correct combined sex age height and weight values) is because of the condition IF age < 60.   But the author NEVER explained how he got the age < 60, it just pop up. And if you put any other conditions or no conditions at all, you would get a messed up result with obs having incorrect values.   Eg, no IF condition proc sort data=spg.patients;by id; run; data one2one; set spg.patients; set spg.measure; run; proc print data=one2one;run; you get  you can see the obs from ID 8045 are all messed up.     Or if you change from age < 60 to age < 50, you'll get.. still messed up.   So I don't get the point or use of OTOR in real life. When would you want to have the second data set to over write variables in the first data set when combined them and get all messed up observations.   I have attached two SAS files for the data sets if you wish to replicate the results.       ... View more