Is it better for the kids if we stay together?
The honest answer is that it depends on what the children are currently living in, and the research is clearer about that than about marriage itself. Broadly, children in homes with sustained hostility do about as badly as, and on some measures worse than, children whose parents separate. The advantage of an intact household largely disappears once the household is hostile. Children in homes that are unhappy but not hostile tend to find a divorce harder.
It matters which of those is more common, and it is not the one people assume. Researchers estimate roughly half of divorces, by some estimates a majority, come out of low-conflict rather than high-conflict marriages, the branch where children fare worse, largely because the divorce arrives without warning. If you are reading this because the marriage is unhappy rather than hostile, that is most likely the branch you are in.
That is genuinely two different findings, and which one applies to you is not something a web page can tell you.
What is reasonably well established
Conflict matters at least as much as household structure, and possibly more. Children exposed to frequent hostility between parents, especially conflict involving or about them, show worse outcomes on most measures. That holds whether the parents are married or divorced, which is the finding people skip. The associations are modest, though, and researchers are careful not to claim conflict outweighs structure outright. Both appear to matter independently.
Divorce is a real loss, and most children find it hard for a period. Most also adjust, and years later the majority score in the normal range. That is not quite the same as no lasting harm: plenty of them still carry painful memories of the divorce into adulthood, and researchers treat that as grief rather than a sign something is clinically wrong with a child.
What separates the ones who struggle is fairly consistent: continued conflict after the split, being used as a messenger or an informant, losing a relationship with a parent, financial instability, and repeated disruption. Almost every one of those is a decision the adults make afterward.
Which produces an uncomfortable and only partly evidenced summary: children generally do better out of a sustained-hostility marriage than in it, and handling a divorce well helps, but less than people hope. The one study that tested the "good divorce" directly found cooperative co-parenting improved behavior problems and closeness to fathers, and made no measurable difference to grades, self-esteem, substance use, or life satisfaction. And nothing in this literature compares divorce to a marriage that was successfully repaired; that comparison has not been studied. Neither half is a recommendation about your family.
The question that is actually answerable
Not "should we stay together." Try instead:
What are the children living with right now? Not what you intend, what they observe. Children register tension with startling accuracy, including the version parents believe is well hidden. The house where nobody argues but nobody speaks is not neutral to a child.
Would separating actually reduce the conflict, or relocate it? This is the pivot. Parents who separate and keep fighting through the children have given them the loss without the relief. Parents who separate and become civil have usually given them something better than they had.
Is the marriage actually over? A separate question with its own set of things worth sitting with.
Has repair genuinely been attempted? Counseling that both of you engaged with, not one person suggesting it and the other agreeing to stop the conversation.
What are you modeling? Children learn what a relationship looks like from the one in front of them, and that lands either way.
What does not help
Waiting for a milestone. Until they finish school, until the youngest leaves. Sometimes right, often a way of not deciding, and children frequently report having known for years and having watched their parents wait.
Telling them you stayed for them. It hands a child responsibility for their parents' unhappiness, and adult children hearing it later commonly describe it as the worst part.
Assuming they do not know. They almost always do.
If there is any question of safety
Then this is not the calculation. Children exposed to violence or coercion are harmed by exposure, and staying does not protect them from it. That situation has its own help, free and confidential at any hour, and an advocate can think through timing with you. The safety resources page has the numbers.
What is genuinely in your control
Whichever way you decide, the factors that most affect how children come through are the ones you keep hold of: keeping conflict away from them, keeping them out of the middle, protecting both relationships, and holding their routines steady. What goes in a parenting plan is largely a list of ways to make those automatic rather than dependent on goodwill.
This summarizes a large and genuinely mixed body of research, and it is not a prediction about your family or a recommendation either way. Divorcery exists to help people through a divorce, not to talk anyone into one. A family therapist can look at your actual situation, which no article can.