A density plot is a smooth curve that shows how the values of one numeric variable are spread out. Instead of grouping data into bars like a histogram, it uses kernel density estimation to draw a continuous shape. The peaks show where values cluster, and the width shows the spread. It is a clear, clean way to see the distribution of your data.

Paste your comma-separated numbers or upload a CSV file and choose the value column. Then click Generate. This free density plot maker draws a smooth curve right in your browser, lets you edit colors, fonts, and bandwidth, and exports the figure as PNG, JPEG, or WebP. No sign-up or software install is needed.

Kernel density estimation is the method behind a density plot. It places a small smooth bump (a kernel) over every data point and adds them all up to form one continuous curve. The result estimates the probability density of the variable. This tool uses a Gaussian kernel, which is the most common choice.

Bandwidth is the setting that controls how smooth the curve is. A small bandwidth hugs the data and can look bumpy, while a large bandwidth is smoother but can hide real features. This tool starts with Silverman's rule as a sensible default and gives you a slider to fine-tune it.

A histogram shows raw counts in bins, while a density plot shows a smooth estimated shape. Density plots are great for comparing groups and do not depend on bin choice, but histograms are better when you need exact counts. Many researchers show a histogram with a density curve on top.

A density curve represents a probability distribution, and all probabilities must add up to 1. So the whole area under the curve is scaled to exactly 1. This is why the y-axis is called density, not count. To get a probability you read the area between two x-values, not the height at one point.

A bimodal density plot has two clear peaks. It usually means your data is a mix of two groups, for example juveniles and adults, or two machines with different settings. Two humps are a signal to split the data and look at each group on its own, because a single average would hide the pattern.

There is no strict rule, but density plots become stable and trustworthy with a few dozen values or more. With fewer than about ten points the curve wobbles and can mislead. For very small samples, a dot plot or the raw numbers are safer.

A good bandwidth is one where the curve looks smooth but still shows real peaks. Silverman's rule gives a solid starting point. If the curve looks noisy, raise the bandwidth multiplier; if it looks too flat and hides detail, lower it. Always report which rule you used.

Yes. If the curve has a long tail to the right, the data is right-skewed (positive skew). A long tail to the left means left-skew. A curve that is roughly a mirror image around its peak is symmetric. This tool also prints a numeric skewness value to confirm what the shape shows.

This tool uses a Gaussian (normal) kernel, which gives a smooth, well-behaved curve and is the standard choice in most statistics software. The Gaussian kernel weights nearby points more heavily than distant ones, so the density falls off gently on each side of every data point.

The y-axis shows density, which measures how tightly values cluster at each x-value. It is not a count or a percentage. A higher curve means more data is packed into that region. To turn density into a probability, measure the area under the curve between two x-values.

Yes, comparing groups is one of the best uses of density plots. Because each curve is scaled to an area of 1, groups of different sizes can be compared fairly on the same axis. Overlapping curves suggest similar distributions, while separated peaks suggest a real difference worth testing.

They are closely related. A probability density function (PDF) is the true, theoretical curve for a variable. A density plot is an estimate of that PDF built from your sample data using KDE. With more data, the estimated density plot gets closer to the true PDF.

Silverman's rule is a simple formula for choosing a good bandwidth automatically. It uses the standard deviation, the interquartile range, and the sample size to balance smoothness and detail. It works well for roughly bell-shaped data and is a trusted default in many statistics packages.

Yes. Once the figure looks right, click PNG, JPEG, or WebP to download it. You can set the export width and height for high resolution, and the figure caption you type is baked into the image, so it is ready to drop straight into a manuscript or slide.

A Gaussian KDE can spread a little past the smallest and largest data values, because each kernel has tails. If your variable cannot be negative, this is a display effect called boundary bias, not real data. Lower the bandwidth or trim the axis range to reduce it.

Yes. The tool accepts any real numbers, including decimals and negatives. It simply ignores blank cells and non-numeric text. So temperatures, z-scores, log values, and measurements with decimals all work fine.

Describe the shape in words, give the mean, median, and standard deviation, and reference the figure. For example: "Scores were unimodal and slightly right-skewed (M = 48.2, Mdn = 47.1, SD = 9.4); see Figure 1." State the kernel and bandwidth rule in your methods so the figure is reproducible.

Yes, it is completely free with no sign-up, no watermark, and no install. Your data stays in your browser and is never uploaded to a server. You can generate, restyle, and export as many density plots as you like.