Part 1 of the summary can be found here.
This study was conducted by Rich Ling in the early 2000's, thus some of the phones described have older features.
The messages were analysed by their type/token ratio, defined as the number of different words (type) compared to the total number of words (token). The researcher's hypothesis is that the faster a medium of producing language becomes, the lower the type/token ratio would be. The researcher also justifies the hypothesis, saying that writing by hand, or even writing for formal purposes takes a long time, and one would be likely to draw on a wider range of words when there is a longer delay between thought and expressing said thought.
However, based on the type/token ratio of the studied SMS messages, it appears that SMS is an unusual case. The range of type/token ratios mentioned in the article is between 0.487 and 0.609. SMS messages are typed in with the keypad of one's cell phone, which can take a longer time to produce coherent messages.
According to the hypothesis stated in the previous paragraph, a slower method of writing should give more time for the user to think of different words, leading to a higher type/token ratio. The results in the article show otherwise, having an average type/token ratio compared to several reference texts.
The researcher points out this discrepancy by stating that SMS users often have to focus on the writing tool, that is, the keypads of their phones, so they are less likely to think of unusual words. Users between the ages of 16 to 19 years of age have the lowest type/token ratio, whereas users between the ages of 35 and 44 have the highest type/token ratio.
When the messages were analysed by mean word and message length as well as complexity, there were no statistically significant differences in the length of words between genders or age groups. There was also no significant age-based difference in the mean number of words per message. There were gender-based differences in SMS message complexity and number of words in a message.
Women were more likely to write longer and more complex SMS messages, with 6.95 words per message on average. In contrast, men tend to write simpler and shorter messages, with an average of 5.54 words per message. Females in the 16-19 age group were found to write the highest amount of complex SMS messages, while males in the same age group wrote the simplest messages.
Surprisingly enough, abbreviations weren't used often in the sampled SMS messages, and any abbreviations that showed up were mainly used by teenagers and young adults. Users who were between 20 and 24 years of age were also more likely to capitalize and punctuate their messages properly. Besides, women or younger users wrote most of the messages that included salutations or closing phrases.
The article concludes that SMS messages have features of both spoken and written language with their own unique characteristics. SMS messages are similar to spoken language in terms of immediate communication and informality. SMS is similar to writing in the assumption that the participants aren't in close proximity, and the text of an SMS message can be edited before the message is sent.
The unique characteristics of SMS messages arise from the electronic medium they originated in, such as the ability to save messages in one's cell phone. These messages only last as long as the phone's memory remains intact, which may not be as long as a letter's storage period. Norwegian women are more likely to send complex SMS messages, express their feelings and plan for more immediate events. Young adults are more likely to write messages that contain advanced capitalization and punctuation.