Secure multiparty computation (MPC) enables multiple parties to jointly evaluate functions while ensuring the privacy of their inputs. Sampling a biased coin is an important MPC building block for evaluating randomized functions. This paper presents a new MPC protocol for sampling a biased coin using \(2d+1\) unbiased coins. The protocol is statistically secure against passive adversaries and can be implemented using \(11d + 5\) multiplications and five rounds. Here, d is associated with the used finite field size p as \(\lceil \log _2 p \rceil = 2d + 1\) . The protocol is based on secure arithmetic in \(\mathbb {Z}_p\) and can be implemented using any linear secret-sharing scheme. Active security for this protocol can be achieved by incorporating additional existing protocols. The proposed protocol offers significant reductions in round and communication complexities compared to a solution offered by Eriguchi et al. that requires \((5n + 19)d\) multiplications and 11 rounds for n parties.