Some people have a mistaken idea of what a scientific theory is, and that misunderstanding (often coupled with overly enthusiastic writers for the mass media) can cause them to be swayed by propaganda.
Science has, as a field, learned to respect the limits of empirical knowledge. It views progress in the sciences as the replacement of earlier attempts to explain some phenomenon with more refined explanations when experience shows the old explanations to have flaws. So science both expects to be found wrong and also looks forward to being found wrong because the eventual result will be a more comprehensive understanding. It is a little bit like Churchill's view of American government in action. He claimed that the Americans kept trying things that did not work and then abandoning each of them in favor of something that had been determined to work better.
A good example of the fall of one scientific theory and the rise of others to take its place is Newtonian Physics. Newtonian or classical physics is so good that toward the end of the 19th century many scientists believed that humans were on the verge of complete knowledge of the physical world. Indeed, classical physics is so good that it is still perfectly adequate for most of the calculations used in industry today. Within certain limits, it is still believed to be entirely correct. But those several problems left over at the end of the 19th century would not submit quietly. Instead, they forced two revolutions in physics: Relativity and Quantum Physics.
The following story has long been used in the philosophy of science to explain why scientists do not claim to have proven anything. The study of ornithology has one minor branch that is concerned with swans. One of the simple questions about swans is: "What color are they?" One could start in the far north eastern part of Europe and examine swans. The swans indigenous to these places would, one after another, be recorded as white. Probably many people to this day would affirm, "Swans are white." But very soon after explorers reached Australia they discovered that the swans there are black. The lesson is that one never knows what the next observation or the next experiment will show.
Newtonian physics has been similarly verified over and over again. Nobody has ever found a real exception to it in any of the millions of high school and college physics lab experiments. Even attempts to confirm the numbers with very highly accurate instruments showed that the agreement of experimental data with theory just got closer and closer. Everything, that is, until people examined what happens when objects are moving very fast (i.e., at some substantial fraction of the speed of light) or when objects on the scale of atoms and even electrons are examined. Suddenly it was as though humans had entered a different reality.
At extremely high speeds, time does not pass at the same rate. If one puts a clock on a satellite and an identical clock is left on earth after the two are synchronized, the speeding clock will be found to lag the earthbound clock by a calculable amount. The math for determining the lag is not even hard. The formula that describes the lag can be derived using high school algebra. So classical physics, the one that works so well on the ground, will not work even on something so close to earth as the GPS system. The reason is that the satellites used for global positioning are moving enough faster than are humans on earth to make a practical difference.
At extremely small scales of operation, things do not work as expected either. A single unit of light (a photon) is emitted from one place. There are special laser devices that can emit single photons on demand. When the unit of light hits a distant piece of photographic film, it hits in one place and all the energy it left the laser with ends up being transferred to an atom in the emulsion. But if a barrier screen is put between the laser and the detection screen and the barrier screen is pierced by two narrow slits that are very close to each other, then this supposed "particle" will go through both holes and then it will interfere with itself in the way that noise reduction headphones can cancel out (or potentially augment) outside noise by creating an inverted version of the noisy sounds. If there are three slits, the unit of light will go through all three slits. Almost all the time the unit of light will show up on the detection screen, but sometimes it shows up on the other side of the detection screen. (That's called "quantum tunneling" and is important in some electronic chips or similar devices.)
People in the general population may have trouble believing in science because they have been led to equate "well grounded" (meaning, essentially, that the predictions of theory have been checked experimentally over and over again) with "proven."
On top of that fundamental misunderstanding, secondary sources frequently display too much enthusiasm for the latest scientific studies and claim that something has been "proven" when all that has actually happened is that somebody has gotten some good results in some line of inquiry and is very hopeful that it will continue to pan out over the long haul. Then when the next swan is examined it turns up black and the scientists have to revise or abandon their hopeful theory.
Generally speaking, creating a scientific understanding of something is rather like assembling a huge jigsaw puzzle. With no picture of the puzzle solution at hand, people try to fit pieces together and they get several unrelated patches organized, but they may guess wrong as to how those isolated patches are going to fit together in the long run. The one thing in some study such as archeology is that the picture is just going to fit closer and closer together, with fewer gaps left over to fill in the imagination, as time goes on.
A theory is often discussed not as a representation of the true state of things but as a model that works well enough to predict how the real thing works to be useful to us for planning purposes. Models are things that human minds make up to try to mimic the workings of the real world in ways that can be manipulated for making predictions. So they are not actually right or wrong, they are merely more or less useful. Newton's model has proven so useful that we don't throw it away. We just pull out the more sophisticated models when the situation demands.