Let $\xi_1,...,\xi_n$ be independent and equally distributed Bernoulli random variables. Also let $\mathbb{P}(\xi_i=1)=p,$ $\mathbb{P}(\xi_i=-1)=1-p$ where $i=1,...,n.$ Prove that $$\mathbb{P} \Bigg( \Big| \frac{\xi_1+\xi_2+...+\xi_n}{n}-(2p-1)\geq \varepsilon \Big| \Bigg) \leq 2 e^{\frac{-n\varepsilon^2}{4}}.$$
So I think I should use Chebyshev's inequality $\mathbb{P}(\xi\geq\varepsilon)\leq \frac{\mathbb{E}\xi}{\varepsilon}$. But I don't know how to start. How to begin this proving?