Bar charts are frequently used to depict mean values. Previous work revealed two interpretation problems. The within-the-bar bias (Newman & Scholl, 2012) refers to the assumption that a value below the mean (inside the bar) is more likely to be part of the underlying data distribution than a value above the mean (outside the bar). The bar-tip limit error (Kerns & Wilmer, 2021) refers to the assumption that the bar-tip is the upper limit of the data distribution, as if the bar chart showed a count. In the present work, it was investigated if the within-the-bar bias is only a product of the bar-tip-limit error or if it can occur independently. An online experiment was conducted with the between-subject factors data point (inside the bar vs outside the bar) and explanation (not given vs given). The explanation demonstrated that the data points can appear both inside and outside the bar, thus indicating that the bar-tip is not the limit. The results showed a replication of the within-the-bar bias, as participants rated the likelihood of the point inside the bar as higher than the point outside the bar. Importantly, the within-the-bar bias remained when all participants who conducted a bar-tip limit error were excluded. In line with this pattern, the explanation only led to a small reduction of the within-the-bar bias. Taken together, the data suggests that the within-the-bar bias can occur independently from the bar-tip limit error.