SSPS is really design for large samples, like if you had a questionnaire with twenty range items and like 5,000+ respondents. I know it's a lot easier to use for that sort of thing, like when you need to correlate large data sets with multiple variables being used to determine one thing. My wife used it to correlate global and specific just world belief with family structure. She had like 10 questions which had to be analyzed to predict the respondents belief in a just world, and how that just world belief applied to them and to others, then correlate it to things like birth order, siblings with disabilities, parent's marital status, etc. You could probably do this in Excel, but it's a lot easier in SSPS and if it were a "real" research project with several thousand respondents, I'm pretty sure excel would get pretty bogged down.
Though, for this, with only two variables, Excel would work more than fine.