A particular form of Reality Is Unrealistic wherein a particular work is shaping the ways that crime is perpetrated, discovered, and prosecuted. It shows up in a number of ways:
The only way to prove anything in a criminal case beyond reasonable doubt is to use the most state-of-the-art, comprehensive analysis.
No, this isn't true, but in some stories it is a Self-Fulfilling Prophecy.
It was first reported in the news media in 2005. The Trope Namer is CSI; it, its spinoffs, and its imitators have led juries to refuse to consider any evidence trustworthy except for the most comprehensive, and to equate its absence with reasonable doubt. Whether or not this is actually true, or that CSI caused this if it is, doesn't matter to us. It is what the trope is called, that's all. Interestingly, the Season 15 opener of CSI was called "The CSI Effect"!
Collecting this super-evidence takes much less time In-Universe in CSI-like stories than it does in Real Life because of time compression — both to make the show fit in an hour with ads and to let a season take less than a few years In-Universe. In the show, DNA tests work in hours; in Real Life, it can take months (a DNA analysis can take a few days to complete and American crime labs are infamously overworked and backlogged). Unfortunately, jurors are frequently unaware of the time difference; getting evidence beyond a reasonable doubt by the CSI measure can conflict with giving the defendant a fair and speedy trial. And heaven help the prosecutor if the jury wants evidence of a kind that doesn't exist yet! At least one of the CSIs has had security camera footage rotated to see the back of something, for instance.
Most of the trouble comes from forgetting the "reasonable" part. You don't need ironclad proof of someone's guilt to convict them; you just need enough that there is no credible scenario in which they're innocent. For example, it's possible that Defendant X didn't mean it when he said he wanted the victim dead, picked up the murder weapon just to look at it, left his DNA at the crime scene before the murder, and Googled "body disposal" out of morbid curiosity. But all that's a little much to ask a jury to swallow.
On the plus side, some say jurors who expect forensic evidence are less likely to place excessive trust in the accuracy of witness testimony, and that this may reduce the risk of wrongful convictions. Witness testimony has been thoroughly demonstrated to be one of the least reliable forms of evidence and one of the most frequent causes of wrongful convictions. See the Real Life sections of Fake Memories, Laser-Guided Amnesia, and Weirdness Censor for just a few reasons why some groups advocate for the elimnation of testimony as evidence altogether in favor of purely forensic-based trials.
Criminals who are aware of crime scene analysis may take steps to attempt to throw off the analysis.
These steps are usually relatively simple, such as picking up spent shell casings, or wearing gloves. Some criminals have deliberately planted DNA samples from other plausible suspects to derail any investigation. In one particularly extreme case, Jermaine McKinney of Ohio committed a double murder, then cleaned the crime scene (and himself) with bleach to destroy DNA traces, bundled the bodies before transporting them and lined his car trunk with plastic to prevent fiber and blood contamination, collected his cigarette butts and took numerous other steps; he was only caught because he was unable to dispose of his murder weapon (a crowbar) in an unexpectedly frozen lake and simply left it sitting on the ice.
Note that some law enforcement officials believe this can be a good thing, as many criminals end up generating more evidence by trying to dispose of it. Max Houck, director of the Forensic Science Initiative at West Virginia University, gave the example of a man who would not lick his envelopes to avoid leaving DNA, only to leave hair and fingerprints in the adhesive tape he'd use to seal the envelopes.
Civilians who are aware of crime scene analysis leave crime scenes alone or take steps to preserve the evidence.
This helps make up for the second effect.
One British kid, remembering something he'd seen in CSI: Miami, managed to preserve some evidence in a case — and got in the papers for it.
This phenomenon predates CSI by quite a bit. For example, during the Manson murder case, a kid found a gun in a field. Having watched Dragnet, he knew to pick it up by the muzzle to preserve fingerprints. Unfortunately the cops didn't, and grabbed the gun by the handle, obliterating any fingerprints.
Some CSI labs have complained that more police officers call them to crime scenes and submit much more evidence to them. Yes, that's right; real-life CSI techs are complaining about being expected to help solve more crimes. This is because real life CSI labs these days tend to have a month or two of backlog before they take on new cases. Making them take relatively minor cases hurts more than it helps.
Forensic reports from the prosecution are always considered true.
This one, when it happens, severely annoys defence attorneys.note This happens even though the science behind forensics is constantly evolving, and many labs are using methods no longer considered certain.
One notorious case down in Texas involved a man who was convicted on the basis of his lip prints — a method of identification that hasn't been widely considered reliable for decades. Even worse, a man in Oklahoma spent almost two decades on death row for a crime that it was later proved he didn't commit. The primary evidence against him, aside from blatantly specious interrogation tactics? The shape of his hair follicles, a mode of identification that was proven useless in 1885, over a century before the murder took place.
Another notorious case in Texas has been the actual execution of a man accused of setting a house fire which killed his children. This one is very touchy for people on both sides of the issue of the death penalty for obvious reasons, but one fact which is beyond dispute (amongst experts not directly involved in the prosecution) is that the investigator used apparently clear signs of arson that research since the original trial has been demonstrated to not be ironclad proof of arson at all.
The arson-evidence controversy later was explored in the Law & Order: SVU episode "Torch," portraying two fire scene investigators arriving at opposing conclusions about whether a man had set fire to his house, killing his family. The first investigator, near retirement, relied on outdated methods of interpretation of the evidence; the second investigator (with the help of a District Attorney played by Sharon Stone), recreated the scene and demonstrated that the other investigator's interpretation was flawed. The episode is discussed in this news article.
See 24 for a similar phenomenon. Compare The Perry Mason Method, which inspired a similar problem in the 60's where some juries wouldn't convict unless the accused confessed in court. Wikipedia has a featured article on this topic. Of course, Wikipedia itself is a focus of the dangerously strong CSI effect.