Step 1: Recall the definition and range of the correlation coefficient:
The correlation coefficient, usually written as r, measures how strongly two numerical variables move together in a straight line relationship.
By definition, it is a normalised quantity, and this normalisation forces it to always lie between -1 and 1.
\[ -1 \le r \le 1 \]
Step 2: Understand what the extreme and middle values mean:
When r equals 1, the two variables increase together perfectly, that is a perfect positive linear relationship.
When r equals -1, one variable increases exactly as the other decreases, a perfect negative linear relationship.
When r equals 0, there is no linear relationship at all between the two variables.
No real value of r can ever step outside this closed interval, no matter how the data is arranged.
Step 3: Test each option against this range:
Option (A), 0, lies right in the middle of the interval, so it is a valid value.
Option (B), 10.3, is far greater than 1, so this can never be a correlation coefficient.
Option (C), -0.9, lies between -1 and 1, close to the negative end, so it is a valid value.
Option (D), 2, is greater than 1, so this is also impossible.
Final Answer:
Only values that stay within -1 and 1 can be correlation coefficients, so 0 and -0.9 qualify.
\[ \boxed{0 \text{ and } -0.9} \]