Keyboarding Performance and Analysis

Typing Speed Measures Introduction Understanding typing speed and text-entry quality is essential for evaluating human performance with keyboards. Whether assessing a dispatcher's job readiness, comparing different keyboard designs, or even verifying someone's identity through their typing patterns, we need reliable ways to measure and compare typing behavior. This section explores the standard metrics used to quantify typing performance and the research tools that help us understand the typing process in detail. Words Per Minute (WPM) Words per minute, or WPM, is the standard metric used to measure typing speed in recruitment, competitions, and general performance assessment. It provides a simple, comparable number that lets us talk about typing speed across different contexts. The key to WPM's usefulness is its standardized definition of what counts as a "word." In typing measurement, one word equals exactly five characters or keystrokes, regardless of how many actual letters the word contains. This means: "Brown" (5 characters) = 1 word "Mozzarella" (10 characters) = 2 words "A" (1 character) = 0.2 words This standardization is crucial. Without it, comparing typing speeds would be nearly impossible—someone typing technical documents with long words couldn't be fairly compared to someone transcribing simple text with short words. By converting everything to this standard measure, WPM allows meaningful comparisons across different languages, documents, and hardware. Alphanumeric Entry Speeds Different groups of users show predictable typing speed ranges. Knowing these benchmarks helps you understand what constitutes fast, average, or professional-level typing performance. Average computer users type at modest speeds: Transcription (copying existing text): approximately 33 WPM Composition (writing original text): approximately 19 WPM Notice that composition is significantly slower than transcription. This makes intuitive sense—when you're copying text, you already know what comes next. When composing, you must think about content while typing, which slows you down considerably. Professional typists and those in typing-intensive jobs maintain much higher speeds: Typical range: 50–80 WPM Job requirement for dispatchers and similar positions: 80–95 WPM minimum Advanced typists can reach elite levels: Speed threshold: exceeding 120 WPM These benchmarks vary by country and profession, but the general progression is consistent: average users are slowest, professionals are much faster, and advanced typists approach speeds that most people find remarkable. Text-Entry Research Metrics When researchers study typing, they need more detailed metrics than just WPM to understand what's happening. These research metrics break down the typing process into components, revealing information about both speed and accuracy. Classification of Keystrokes Every keystroke a person makes falls into one of four categories, depending on whether it appears in the final text and whether it was an error: Correct (C): Keystrokes that appear in the final transcribed text and are correct. These are the "good" keystrokes that contribute to the final output. Incorrect Not Fixed (INF): Erroneous keystrokes that were not corrected and remain in the final transcribed text. These are mistakes that made it through to the end—they represent true errors. Fixes (F): Keystrokes used for editing and correction, including: Backspace and delete keys Cursor movement keys Modifier keys used for editing operations These keystrokes don't produce visible characters in the final text; instead, they modify what's already been typed. Incorrect Fixed (IF): Erroneous keystrokes that were caught and removed during editing. Unlike INF errors, these mistakes were corrected before the final text was produced. Understanding these categories matters because they tell us different things about typing performance. Two people might have the same WPM, but one might have made many errors that were carefully corrected (high IF, low INF), while the other made few errors overall (low IF, low INF). The keystroke breakdown reveals these differences. Minimum String Distance Error Rate Minimum string distance (MSD) measures typing accuracy by counting how different the final typed text is from the target text. Specifically, it counts the minimum number of single-character edits needed to transform the typed text into the correct text. These edits include: Insertions: adding a character Deletions: removing a character Substitutions: replacing one character with another For example, if you typed "teh" instead of "the," the minimum string distance would be 1 (one substitution fixes it). The important feature of MSD is this: errors that are corrected during typing do not increase the error rate. If you type "teh," immediately notice the mistake, backspace, and type "the," the final text is correct and the MSD error rate reflects that. This makes MSD measure the accuracy of the final output, not the number of mistakes made along the way. This differs from keystroke-level metrics—you might have made an error and spent several keystrokes fixing it, but MSD only cares about whether the final result is correct. Keystrokes Per Character (KSPC) Keystrokes per character (KSPC) measures efficiency by calculating the average number of keystrokes required to produce one character in the final transcribed text. It's calculated as: $$\text{KSPC} = \frac{\text{Total Keystrokes}}{\text{Characters in Final Text}}$$ A KSPC of 1.0 means perfect efficiency—every keystroke resulted in a character in the final text with no corrections or errors. A KSPC of 1.5 means it took an average of 1.5 keystrokes to produce each final character, suggesting either errors that were corrected or some inefficient input method. Important limitation: KSPC cannot distinguish between corrected and uncorrected errors. A high KSPC might indicate: Many errors that were carefully corrected before finishing, or Many uncorrected errors remaining in the final text Both scenarios produce high KSPC values, but they reflect very different typing quality. You need additional metrics (like INF and MSD) to understand what the KSPC number actually means. When to use KSPC: KSPC is appropriate for comparing similar input methods—for example, comparing two different keyboard layouts, both used with the same physical keyboard. However, KSPC is not suitable for comparing fundamentally different devices, such as a QWERTY keyboard versus a multi-tap mobile phone keypad. The input methods are so different that KSPC values wouldn't be meaningfully comparable. <extrainfo> Keystroke Dynamics Keystroke dynamics, also called typing dynamics, is a different application of keystroke analysis. Rather than measuring speed or accuracy, keystroke dynamics captures the precise timing of each key press and release as a person types on a keyboard. This timing information is used for biometric identification—authenticating someone's identity based on their unique typing pattern, similar to how speaker recognition identifies people by their voice. Just as no two people have identical voices, no two people have identical typing rhythms. By analyzing patterns like the time between keystrokes, how long each key is held down, and the overall cadence of typing, systems can verify that the person typing is who they claim to be. </extrainfo>