The term "ceiling effect" has two different meanings in social science research.
First, a ceiling effect is seen when an independent variable no longer affects a dependent variable after the independent variable reaches some particular level. As an example, let's say that the richer a person becomes, the more likely he or she is to be a Republican. But let's also say that once the person's wealth (the independent variable) reaches a certain level, getting more money (increase in the independent variable) does not make the person more likely to be a Republican (no more effect on the dependent variable.
Second, "ceiling effect" can refer to studies in which a variable is not measured above a certain level. For example, a survey of how much money people make might have a last category for $1 million per year and up. This is a ceiling effect because the variable (income) is no longer measured past the $1 million per year level.