* The original number of students: Removing 3 students from a class of 10 will have a much larger impact than removing 3 students from a class of 100.
* The original distribution: Was it roughly normal, skewed left, skewed right, uniform, bimodal, etc.? Knowing the original shape is crucial.
* The exact scores: We need to know the precise scores of all students to accurately determine the impact of removing those 3 low scores.
Possible Scenarios:
* Original distribution was roughly normal: Removing 3 low scores might make the distribution slightly skewed to the right, but it's likely to remain roughly symmetrical.
* Original distribution was skewed right: Removing 3 low scores would likely reduce the right skew, potentially making it more symmetrical or even slightly skewed left.
* Original distribution was skewed left: Removing 3 low scores would have little impact, as the skew would likely remain.
In short: Without more information, it's impossible to determine the new shape of the distribution. Provide the original data or a description of the original distribution, and I can help you analyze the effect of removing those 3 students.