The idea with Levene's test is to see if the two groups you are testing have
distributions that differ significantly. If they differ you take the lower row of significance and degrees of freedom. The degrees of freedom are smaller and has decimals, which normally degrees of freedom never have.
My groups had 36 and 10 persons, t was
2,256, df was 16,88 and p=,038. This tells you that Levene's test was significant.