Qwerty to Dvorak, A PhD thesis, 1978

By Xah Lee. Date: .

Keyboard Retraining: Qwerty to Dvorak

A PhD thesis by Bradley John Lessley, 1978

Source: https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/8336h4796

Xah's Note

This article is not of much value today, except historical.

In a very hard-to-read and wordy way, it tries to answer the question, about whether for people already know how to touch-type QWERTY, how long would it take them to learn Dvorak, and if their speed would improve.

The conclusion of the article is that, after 100 hours of retraining, yes, the speed improvement may be 12 to 17 percent. (see page 66.)

Note, if you train 1 hour per day, then 100 hours would be 3 months.

The article has significant historical value though. It was written in 1978.

to give a sense of what typing is like in 1978, here's 2 popular computers at the time.

terminal ADM 3A
ADM-3A, 1976
DEC VT100 terminal 2013 08 30 7ea01a58 s s265x236
DEC VT100, 1978

Note, the screen can only show 80 x 40 characters.

The article, gave summarization of about 8 previous research studies. Also, their conclusion is similar to the article.



The purpose of this study was to provide information pertaining to performance levels of Dvorak keyboard operators through retraining. More specifically, the purposes of this study were:

1. To compare operator performance rates on the Dvorak keyboard with operator performance rates on the Qwerty keyboard.

2. To determine whether keyboard operators can learn to use the Dvorak keyboard if they were initially trained to use the Qwerty keyboard.


The participants in this research study included 16 keyboard operators employed in public and private sector agencies and organizations in Oregon. Eight operators used CRT word processors and eight used element correcting typewriters. Each participant received ten hours of keyboard introduction instruction on the Dvorak keyboard, followed by 100 hours of skill-development keyboarding involving both skill-development drill exercise and daily office production correspondence.

Nine null hypotheses were developed regarding Dvorak keyboard retraining rates.

The t test and one- and two-factor analysis of variance with repeated measures were used to test these null hypotheses. Trend analysis was also used to describe the data.


1. Dvorak keyboard production and straight-copy rates, after 100 hours of skill-development keyboarding, were not significantly greater than preretraining Qwerty keyboard production and straight-copy rates for both CRT and element keyboarding equipment. The learning curve slopes, however, were still upward at the end of 100 hours of skill-development keyboarding, suggesting that production and straight-copy learning would continue on both types of keyboarding equipment.

2. Dvorak keyboard production rates were found to be significantly different at 20- and 25-hour intervals for CRT and element keyboarding equipment, respectively, for 100 hours of skill-development keyboarding. The same was found for combined CRT/element Dvorak keyboard straight-copy rates at ten-hour intervals.

3. The Dvorak keyboard production rate of learning on both types of equipment was found to be linear for 100 hours of skill-development keyboarding.

The Dvorak keyboard straight-copy rate was found to be curvilinear. Production rates were not combined as they were for straight-copy performance, since different performance measures were used with each type of keyboard equipment.


1. Studies should be undertaken involving retraining of students in a classroom environment.

2. Studies should be undertaken to determine production performance unique to certain keyboarding tasks. For example, performance indicators should be established for transcription and rough-draft production activities.

3. Research projects should be undertaken addressing reduction of operator fatigue and ease of keyboard operation resulting from retraining on the Dvorak keyboard.

4. Studies should be undertaken to compare the learning rates of beginning keyboard students on the Dvorak and Qwerty keyboards under experimental conditions.

Keyboard Layouts

  1. qwerty heatmap  to build a fire-s365x171
    Alternative Layouts
  2. Alt Graph, Compose, Dead Key
  3. Most Efficient Layout?
  4. Maltron vs Dvorak
  5. Colemak vs Workman


  1. Dvorak Layout
  2. Hardware vs Software Dvorak
  3. Myth of QWERTY vs Dvorak
  4. Dvorak vs Colemak
  5. Blank Keycaps vs Labeled Keys
  6. List of Dvorak Keyboards
  7. Qwerty to Dvorak, A PhD thesis, 1978


  1. International Layouts
  3. German
  4. German Ergonomic
  5. French
  6. New French Layout
  7. French Ergonomic
  8. French Letter Frequency
  9. Russian
  10. Portuguese Ergonomic
  11. chinese pinyin input 2017-07-19
    Chinese Input Methods
  12. Pinyin Letter Frequency 拼音字母頻率
  13. Japanese input method menu 39864
    Japanese Input Method

Layout Efficiency

  1. English letter frequency To Build a Fire  Jack London 50383
    Char Frequency Counter
  2. Most Efficient Layout for Numbers?
  3. Semicolon vs Apostrophe Frequency
  4. Inverted Number Row Better?

If you have a question, put $5 at patreon and message me.