Peer-reviewed research
KSPC (Keystrokes per Character) as a Characteristic of Text Entry Techniques
I. Scott MacKenzie, Proceedings of the Fourth International Symposium on Human Computer Interaction with Mobile Devices (Mobile HCI 2002), LNCS 2411, Springer, 2002 · DOI 10.1007/3-540-45756-9_16
Read the originalRead on October 16, 2026
- What it is
- Defines keystrokes per character (KSPC) as a measure of text entry techniques and computes it for methods such as multitap, dictionary-based disambiguation and word prediction.
- How much weight it carries
- Peer-reviewed conference paper by a leading text entry researcher, read here in the author's own HTML copy. It describes a design metric rather than measuring people.
Facts taken from this source
- KSPC is the number of keystrokes a text entry technique needs, on average, to generate one character of text in a given language.
“KSPC is the number of keystrokes, on average, to generate each character of text in a given language using a given text entry technique.” (Abstract)
- For English, KSPC values across text entry techniques range from about 10, for methods that use only cursor keys and a select key, down to about 0.5, for word prediction techniques.
“Values for English range from about 10 for methods using only cursor keys and a SELECT key to about 0.5 for word prediction techniques.” (Abstract)
- For the basic QWERTY keyboard, MacKenzie's calculation gives a KSPC of 1.0000, because each letter has its own key and each keystroke produces one character.
“we conclude the following for the basic Qwerty keyboard: KSPC = 1.0000” (Section 3, Qwerty Keyboard (KSPC = 1))
- KSPC is the average number of keystrokes needed to produce each character of text with a given text entry technique.
“KSPC is the number of keystrokes, on average, to generate each character of text in a given language using a given text entry technique.” (Abstract)
- For English, KSPC runs from about 10 for methods that use only cursor keys and a select key to about 0.5 for word prediction.
“Values for English range from about 10 for methods using only cursor keys and a SELECT key to about 0.5 for word prediction techniques.” (Abstract)