Ironically, pretty much the only subfield of computer science using the scientific method in any significant capacity is software engineering.
Historically and foundationally CS is a child of mathematics and electrical engineering, neither of which really fit the narrow English term of science.
CS is mostly preoccupied with a) making (useful) deductions in closed logical systems, and b) building cool shit. Empirical knowledge generation is usually only relevant as a tool to help us with the first two goals.
Yes, a lot of CS is mathematics (and EE), but it is not just that.
There is a notion of experimentation in pure CS, because CS asks the question, what can be computed on physical machines. And to check that, you actually need to build and check.
Nothing illustrates this better than Quantum Computing. Computer Science (and Physics) theories predict the existence of quantum computing, and suggest a computational advantage, but we will never know unless we build QCs and experiment.
Historically and foundationally CS is a child of mathematics and electrical engineering, neither of which really fit the narrow English term of science.
CS is mostly preoccupied with a) making (useful) deductions in closed logical systems, and b) building cool shit. Empirical knowledge generation is usually only relevant as a tool to help us with the first two goals.