A researcher is testing a null hypothesis $H_0: \mu = 100$ against a two-tailed alternative. If the observed sample mean is $103$ with a standard error of $1.2$, and the test yields a $p$-value of $0.012$, what decision is appropriate at a significance level of $\alpha = 0.01$?