How to use the csv statistics viewer
- Paste your CSV text into the box, or choose a .csv or .tsv file. Press Use an example to see how it works.
- The tool detects the separator from the first line. Choose comma, semicolon, tab or pipe yourself if the guess is wrong. Untick First row has column names if your file starts with data.
- Read the summary: the number of data rows and columns, then one table for the number columns and one for the text columns.
- Use Download CSV to save the statistics, or Copy tables to paste them into a message or a spreadsheet.
A small order list
A file with the columns Item, Region, Units and Price and five data rows has two text columns and two number columns. Units has the filled cells 12, 1,200, 7 and 30, with one empty cell, so it has 4 filled and 1 empty, a sum of 1,249, a mean of 312.25, a median of 21, a minimum of 7 and a maximum of 1,200. The sample standard deviation is 591.915746.
Price has $1.50, $4.25, $12.00, $1.50 and (2.00), where the brackets mean a negative amount. That gives a sum of 17.25, a mean of 3.45, a median of 1.5, a minimum of -2 and a maximum of 12. Item has 4 different values, with Pen the most common, twice.
What counts as a number column
A column is treated as numbers when every filled cell is a number. A number may have a plus or minus sign, a dollar sign, commas that group thousands in the right places, decimals, or brackets for a negative amount. Percent signs, units and text make a cell not a number. If even one filled cell is text, the whole column is described as text, with a count of how many of its cells were numbers so you can spot the odd one out.
Empty cells, and cells holding only spaces, are counted as empty and left out of every statistic. The count of filled cells is shown beside the empty count so you can see how complete each column is.
How each figure is worked out
The sum is added with compensation so that long columns do not pick up rounding error. The mean is the sum divided by the filled count, and the median is the middle value, or the average of the two middle values when the count is even. The standard deviation is the sample version, which divides by one less than the count, and is blank for a single value.
For text columns the tool shows the number of different values, the most common value and how often it appears, with the first one winning a tie, and the shortest and longest lengths in characters. To go further with a set of numbers, use the descriptive statistics calculator. To check the layout of a file first, try the CSV validator, and to combine files, the CSV merge and split tool.
Privacy and limits
The file is read and analyzed in your browser. Nothing is uploaded. Files up to 10 MB are supported. Quoted values with commas, quotes and line breaks inside them are handled the standard way.
Assumptions and limits
- The first line is column names unless you untick the option, and rows shorter than the longest row count as having empty cells.
- Numbers use a period for decimals and commas only for thousands. Values such as 1.234,56 are treated as text.
- The standard deviation divides by n - 1 (the sample version).
- Dates are treated as text, not numbers.
Frequently asked questions
Why is my column shown as text?
At least one filled cell is not a number under the rules above, for example it has a unit, a percent sign or a misplaced comma. The text table shows how many of its cells were numbers.
How are negative numbers in brackets handled?
A value like (2.00) is read as -2, as in accounting reports. A minus sign works too.
What separator does it use?
It detects comma, semicolon, tab or pipe from the first line. You can choose one yourself in the Separator list if the guess is wrong.
Can it handle large files?
Files up to 10 MB are accepted. Very large tables may take a moment to analyze, because everything runs in the browser tab.