Here's how I would do that. Imagine labeling each of the repeated letters with a subscript so you can tell them apart: \(\displaystyle WI_1S_1CON_1S_2I_2N_2\). Then there are 9 distinct letters so 9! ways to arrange them. But two of those 9! arrangements are, for example \(\displaystyle WI_1S_1CON_1S_2I_2N_2\) and \(\displaystyle WI_2S_1CON_1S_2I_1N_2\), exactly the same except that the two "I"s have been swapped. That is, for every one of the 9! ways to arrange WISCONSIN, there is another that has only the "I"s swapped that we don't want to count as "different". To allow for that, divide by 2!= 2, the number of ways to arrange, or swap, the two "I"s. Similarly with the two "S"s and two "N"s. The number of distinct arrangements is \(\displaystyle \frac{9!}{2!2!2!}= \frac{362880}{8}= 45360\).

More generally, if a word has \(\displaystyle m\) letters, \(\displaystyle n_1\) of which are the same, \(\displaystyle n_2\) of which are the same but different from the first, … to \(\displaystyle n_i\) of which are the same, then the number of distinct arrangements is \(\displaystyle \frac{m!}{n_1!n_2!\cdot\cdot\cdot n_i!}\). Notice that, since 1!= 1, we can include "n"s that are equal to 1 requiring that \(\displaystyle n_1+ n_2+ \cdot\cdot\cdot+ n_i= m\).