OMG! Study finds iPhone texting 2x slower than QWERTY

Updated below: A study has determined that iPhone's touchscreen is two times slower than QWERTY phones when it comes to texting.

OMG!

This study was conducted by User Centric, a Chicago-based usability consultant. User Centric concludes:

It took QWERTY users almost twice as long to create the same message on the iPhone as it did on their QWERTY phone. While there was improvement over time, the difference persisted even after using the iPhone for 30 minutes.

Actually, this isn't all that surprising to me. The big question is whether the gap closes over time. In fact, before reading this study I would have guessed iPhone would have been even slower.

As for the methodology here's what User Centric had to say in a statement:

User Centric tested the iPhone's texting features with frequent texters to see how rapidly they could adapt to the iPhone's touch keyboard. All 20 participants sent at least 15 messages per week. Ten participants owned phones with a full QWERTY keypad and 10 owned phones with a numeric keyboard. None of the participants owned an iPhone. Each participant typed six fixed-length text messages on their own phone and six on an iPhone.

Hole number 1: I'd like to see results from folks that have had an iPhone from the launch. In general, User Centric said that iPhone's touch keyboard was inaccurate. Most users corrected errors with a backspace to erase a character. Does this improve over time?

Hole number 2: The study is based on 20 people. No one owned an iPhone and half of respondents had QWERTY phones. The other half had numeric keypads. A larger sample would have been nice.

User Centric also noted that numeric-based "multitap" texting took about the same amount of time. The consultant said:

In contrast to QWERTY users, numeric users used the "multitap" method of entering text messages on their phones. They pressed individual number keys multiple times to get a desired letter or character to appear. Although multitap is inherently inefficient, numeric phone users took nearly as long to create a message on the iPhone as they did on their numeric phones. There was no increase in efficiency despite the iPhone's corrective text approach.

User Centric did come up with some other interesting takeaways in its study:

• Most participants said their fingertips were too large for the iPhone touch keyboard.
• Participants noticed there was no tactile feedback from the iPhone keypad.
• Participants had 11 errors on the iPhone compared to three on their own phone.
• Five out of 20 respondents wanted a stylus.
• Long fingernails could be a problem.

Update: Jen Allen, a user expert specialist at User Centric, had some comments on my post. Here is her reply in full.

1)  Back in July, User Centric conducted a study testing early adopters of the iPhone.  After these early adopters had used their iPhone for a week, they came in for the study:

One of the anecdotal findings from that study was that users reported having trouble texting with the iPhone, which was the motivation for the current study.

2)  Due to space limitations, the press release did not go into great detail on the technical aspects of the testing.  User Centric's test methodology employed a 2x2 mixed factorial design.  Type of phone owned by the participant was varied between subjects.  However, participants typed text messages on both their own phone and the iPhone during the test session.  Message length, phone order, and message order were both controlled.  While the sample size of 20 may seem low, it did provide us with statistically significant results.

The sample was recruited in the Chicago area.  Participants were required to send at least 15 text messages per week.  In most cases, participants sent many more than this, particularly the QWERTY users.

3)  User Centric has tested over 500 participants for various mobile phone studies in 2007 with approximately 2,000 in the last few years. The company will conduct another study on text entry using mobile phones in the coming weeks.  It should be in the press in another month or so, and will hopefully shed more light on some of the remaining questions surrounding these issues.