Lexical Diversity Calculator
Paste a paragraph, a section or a whole draft and press Calculate. You get the number of words, the number of different words, the type-token ratio (TTR) and root TTR, and a table of your most repeated words. A short guide below explains what the scores mean and why you should compare only texts of similar length. Free, no sign-up, and nothing is uploaded.
How to Use the Lexical Diversity Calculator
- Paste your text. Use at least a few hundred words if you can; very short texts score high almost automatically.
- Press Calculate. The results do not update as you type, so press the button again after editing.
- Read the four scores. Total words, unique words, TTR as a percentage, and root TTR.
- Look at the verdict and the repeated words. A plain-language band follows the scores, and the table lists up to ten words that appear at least twice.
What the Scores Mean
Lexical diversity is how varied the vocabulary in a text is. The simplest measure is the type-token ratio: the number of different words (types) divided by the total number of words (tokens).
- TTR = unique words ÷ total words. It runs from near 0 (the same few words repeated) to 1, shown here as a percentage.
- Root TTR = unique words ÷ √(total words), often called Guiraud's index. It falls less steeply as texts get longer, so it is fairer when lengths differ moderately.
The tool gives a plain verdict: under 40% is low variety, 40% to under 60% is typical, and 60% or above is high. These bands are heuristics for passages of a few hundred words. They do not apply to a 20-word sentence or a 50,000-word thesis.
A Worked Example
Paste this 56-word paragraph: Sleep consolidates memory. During deep sleep the brain replays the day's events, strengthening connections that matter and pruning those that do not. Students who sleep well tend to remember more of what they studied, and they solve problems faster. Skipping sleep to study longer often backfires, because the facts that were crammed are never stored properly. It has 46 unique words, so TTR is 82.1% and root TTR is 6.15, which the tool reads as high variety. Sleep is the top word, with four uses.
Now paste the same paragraph twice. Nothing new is said, but the text is 112 words long with still only 46 unique words: TTR drops to 41.1% and root TTR to 4.35. The same writing scores about half as diverse, which shows both that repetition lowers the score and that length alone lowers it.
Why Length Distorts TTR
Every text reuses common words such as the, of and and, and the longer a text runs the more of its words are reuses. TTR therefore falls as a text grows, whatever the writer's vocabulary. A 100-word sample and a 1,000-word sample by the same author will not give the same TTR. Compare scores only between texts of similar length, ideally cut to the same number of words.
Researchers use measures built to reduce this effect, such as MTLD and HD-D (see McCarthy and Jarvis, 2010, in Behavior Research Methods). This tool does not calculate those, so use it for revision and for rough comparison, not as a research-grade measure.
How Words Are Counted
The text is lowercased and split into runs of letters and digits, with an apostrophe kept inside a word, so don't is one word, whether the apostrophe is straight or the curly kind many word processors produce. Accented letters count as letters, so café is one word. Numbers count as words. A hyphen splits a word in two, so well-known is counted as two. There is no lemmatising: walk, walks and walked are three different words.
Which Tool Do You Need?
| If you want… | Use |
|---|---|
| One variety score for a text | This page |
| Every word ranked by count | Word Frequency Counter |
| Words repeated close together | Word Repetition Checker |
| Density percentages for each term | Keyword Density Analyzer |
| A list of each distinct word | Unique Word Extractor |
Frequently Asked Questions
- What is lexical diversity?
- It is a measure of how varied the vocabulary in a text is, meaning how many different words are used compared with how many words there are in total. It is used in writing instruction, language assessment and corpus research.
- What is a good type-token ratio?
- There is no universal target. For a few hundred words of clear expository prose, 40% to 60% is typical and the tool calls it typical variety. Higher scores are common in short texts and lower ones in long texts, so compare like with like.
- What is root TTR?
- Root TTR, or Guiraud's index, is the number of unique words divided by the square root of the total word count. It reduces, but does not remove, the way plain TTR falls as texts lengthen.
- Why is my score lower when I paste more text?
- Because longer texts reuse more words. TTR measures the share of words that are new, and that share shrinks as the text grows. It does not mean your vocabulary got poorer.
- Does the calculator remove stop words or count word forms together?
- No. It counts every word, including the and of, and treats each spelling as separate, so walk and walked are different words. Counting all words is the usual way to report TTR.
- How can I raise the diversity of my writing?
- Look at the most-repeated words table. Where a word repeats without a reason, vary it or rewrite the sentence. Do not swap exact technical terms for synonyms, because repeating a key term often helps the reader.
- Is my text uploaded?
- No. The scores are calculated in your browser and the text is not sent anywhere.