In recent years, cryptocurrencies have been the focus of significant study and financial interest, and examining social media data is one method for forecasting their pricing. Using works indexed in Scopus between 2018 and 2023, this research examines the effect of social media on forecasts of cryptocurrency prices. 52 publications that predicted bitcoin prices using social media data were reviewed systematically. The findings reveal that Twitter, followed by Reddit and StockTwits, is the most popular social media channel for speculating on the future value of cryptocurrencies. Of the many prediction models available, recurrent neural networks are the most prevalent machine learning method. Evaluation criteria are not standardized between research, however accuracy and mean squared error are two of the most prevalent ones. Data quality concerns, data pretreatment obstacles, and overfitting are just a few of the limits and difficulties that this research shows when it comes to utilizing social media data to forecast bitcoin price. To wrap off, we talk about what this work means for the future of cryptocurrency price prediction research and the impact of social media.