Raynald Levesque: An example of a difficult problem that I have to program is basically a stochastic projection of medical prescription drug costs for a population over a 15-year period. I have my initial population—say I’m going to have 75,000 employees. At each age I have number of persons by sex…this is an employee, I have a spouse, I have three kids. I have to project the prescription drug expenses of those people for 15 years. There are many variables involved. This is a complex and difficult job.

King Douglas: Do you find more pleasure out of solving a job of this magnitude than something that wouldn’t take so long?

Raynald Levesque: Yes, I find pleasure in doing a difficult job, although many of the questions that I prefer to answer on the SPSSX list are relatively small challenges that I can solve in 15 minutes.

King Douglas: Describe the way in which you attack a problem such as the one you described.

Raynald Levesque: For a difficult problem, I use pencil and paper. I write only concepts on paper (no syntax). The whole thing would usually fit on one page.

King Douglas: Among experts, the consensus is that it is important to learn and use syntax. Why is that the case when SPSS makes it possible for the user to do most things via point and click?

Raynald Levesque: I agree with the consensus you are talking about. In my opinion, no serious work can be done without syntax. Syntax is efficient, once debugged it is error proof, it is self-documenting, it is easy to share with colleagues—even colleagues using SPSS in different languages. Point and click is a tool to create syntax. It is not difficult to learn syntax, although macros take some time.

King Douglas: What is your view on annotating SPSS files?

Raynald Levesque: First, I think it is important to write code in paragraphs and to summarize the content of the paragraph in a comment. When the comment is important, I set it off with a row of asterisks. If it is less important, I use just a single asterisk. Annotation can save a lot of time if you need to modify your code at a later time.

King Douglas: What SPSS timesavers do you employ most often?…for example, snippets of code you can reuse, DO REPEAT statements, macros, etc.

Raynald Levesque: All of the above, plus customized toolbars…I think this is generally under-used features. I use scripts to assist in daily chores, for instance, to check that there are no errors in the log, find errors if any, export charts and tables from SPO to Excel, etc.

King Douglas: What goals are you working toward at this time in your life?

Raynald Levesque: Starting in 2006, I expect to be heavily involved in major international consulting assignments. Many of these projects will have actuarial as well as data components. I have also developed a major application in Canada that I would like to adapt to the USA market within the next 18-24 months.