@cy_th wrote: I want to calculate the mean of 'score raw'. How can i do it? I tried to save it as an array and do it but since some values are missing I think it wont work.Also since the array should have elements of the same data type I think what I have done is incorrect. I have attached the excel file here. proc import filename="/folders/myfolders/excel_data" dbms = xlsx out= newdata replace; run; data newdata1; set newdata; array sc(2) score_raw score_num; do i=1; if sc(i) = 'missing' or '' then do; sc(i)=0; end; end; run; proc means data=newdata1; var sc run;   In addition to the other comments on reading data as desired you want to think very carefully about changing a missing value to 0 is appropriate before doing any summary statistic calculation. Consider data such as: value 10 missing missing missing 20 missing missing missing   If I calculate a mean for the non-missing values it would be (10+20)/2 = 15. If I replace the missing with 0 then you are looking at (10+0+0+0+20+0+0+0)/8=> 30/8= 3.75. So, which "mean" do you need? Most of the time leaving "missing" as missing is more appropriate in general. Otherwise you are in the role of imputing values and that should be undertaken carefully. ... View more

Certifications are tailored to learning to pass an exam, not how to solve the daily problems you'll be having. For example, if you always are connecting to a data base why spend a lot of time learning to read text files? Or if a lot of your analysis is basic reporting, learning proc means, freq, report and tabulate are more important than other procedures may be. ... View more

 When you run the program you can see that the first data step only produce one row. That is because each input statement reads a new line of the input data. So, in your case: Read the two first values from the first row.  Read a new line of input data and read the two first values of that row. Return to the start of the code and read a new line of input data. Et cetera!    The second data step has a @ in the end of the first input statement. This tells SAS to wait on the first input data row until you tell it otherwise. So, in your case: Read two values from the first row. Then read two more values.  Return to the start of the code and read a new line of input data. Et cetera!    The @ char is useful if you for example have data that has a different format on different rows. So, if the first value is US you need to use DATE9 as informat and if you have the first value of EUR you use the ISO standard YYMMDD10. Or it might be that the data you have spans more than one row, so row 1 in each set has a name and row 2 has an address.  ... View more