The Improved Shellsort algorithm obviously from the results obtained performs better than Shellsort in the worst-case, the best-case and a small size input of the average-case. The strength of this algorithm becomes more appreciated as the size of the input to it increases for the worst-case and best-case scenarios but when input size begins to be higher than twenty its strength diminishes for the average- case. Implementing the two algorithms on a different platform may produce different running time results but the same pattern will of course show. We therefore, conclude that this proposed Improved Shellsort will run faster than Shellsort for the worst-case, best-case and a subset of the average-case.