Wait, no. That's actually wrong.
The scientific method produces useful results. It adds to our body of knowledge. It leads to technological applications that, in general, have increased the length and quality of our lives ... though not without costs and dangers.
But science is not a machine. It's not a self-programmable computer cum assembly line that takes in raw data at one end and spits out illumination at the other. Science is a methodology.
Because science is a methodology, it can be misapplied, or even barely applied. Scientists get emotionally vested in imperfect explanations. They bend data to fit models rather than models to fit data. They draw incorrect inferences. They commit logical fallacies and mathematical mistakes. They build experiments that can't falsify the theory they're testing. They cherry-pick statistics and results that support their arguments, while leaving out those that contraindicate. Their successes can come by accident, or after years of by-guess-and-by-golly. In biology, under rigid controls of light, heat, moisture, food and temperature, the biota will do as it damn well pleases regardless of your theory. The peer review — the gem of the method — can be sloppily or cursorily done (consider the "Bogdanov affair"), letting shoddy research out into publication.
But most of all, since the scientific method has produced such a humonguous body of results, at least 90% of what any one scientist knows is taken on authority. There's simply not time enough in any person's life to re-check everything. So when a highly reliable theory stops producing expected results, it can take years before scientists start checking the theory for subtle errors; the trust in the system produces an institutional inertia.
Because science is, in a sense, an art — a human art — it's vulnerable to all the weaknesses of fallen humanity: ignorance, sloth, greed, envy, fear, and so on. Because science is done by humans and not machines or angels, it can fail, it can err, it can create public dangers, it can be deliberately subverted. That's why, to work, the sciences need men and women with integrity, compassion and well-formed consciences. That's why scientists should not be treated as gods or superhumans, and should not be given unquestioning credence or obedience.
And that's why I get irritated with secularists who treat Science (and the way they speak of it, you might as well capitalize it) as a cross between the Oracle at Delphi and a soda-pop vending machine. To listen to them, you'd never know any scientist had fudged his results, or that the various disciplines have their internal quarrels over competing theories. The myth of Scientism is quicky eclipsing the quirky, messy, mundanely human reality of science, creating a cult-like atmosphere of fervent credulity, and treating the obiter dicta of people like Stephen Hawking as the Voice of God even when they speak outside their spheres of competence.
In fact, if they're not careful, they'll become everything they've always accused Catholics of being: intellectually-crippled sheep placing blind faith in mere humans. If it happens, it'll be because they've convinced themselves it will never happen.
Yes, science has been wildly successful, and therefore I shall always remain respectful of the scientific method. But it's not The Source of All Knowledge. I refuse to be ruled by the priesthood of the lab coat. I refuse to worship at the Temple of the Method. I refuse to bow down before the Golden Calf of Science.