Visualize relationships between two variables by plotting data points on an interactive chart.
Last updated: March 2026
One point per line, separated by comma or space
A scatter plot is a graph that displays the relationship between two numerical variables. Each point on the plot represents one observation, with its position determined by its x (horizontal) and y (vertical) values. Scatter plots are fundamental tools for exploring correlations, trends, and patterns in bivariate data.
By visualizing data as points rather than as numbers in a table, scatter plots reveal patterns that are often hidden in raw data. You can immediately see if two variables move together (positive correlation), move opposite (negative correlation), or have no relationship (no correlation). They also help identify outliers, clusters, and non-linear patterns.
Scatter plots are used across science, engineering, economics, and social sciences to explore data relationships before conducting statistical tests. They complement correlation coefficients and regression analysis by providing visual confirmation of relationships.
Study Hours vs Test Success
No! Correlation is not causation. If x and y are correlated, it could be: x causes y, y causes x, both are caused by something else, or just coincidence. Always investigate the mechanism.
There's no fixed minimum, but more data produces more reliable patterns. With <5 points, patterns emerge due to chance. Aim for at least 20-30 points for meaningful analysis. Small samples can be very misleading.
Outliers can distort visual patterns. Investigate them first: are they measurement errors, data entry mistakes, or genuinely unusual cases? Sometimes removing outliers, sometimes analyzing them separately makes sense.
Yes! Scatter plots handle negative x and y values. This is common in financial data (losses), temperature data (below freezing), and many other domains.
No pattern suggests no relationship. This is valuable information! However, two variables can be related non-linearly (curved patterns). Consider transformations (logarithm, square root) or other analysis methods.
Use correlation coefficient (r) to quantify linear relationships. Use regression analysis to model the relationship mathematically. Scatter plots provide visual confirmation of these statistics.
Related Tools