Syntactic variables in pupils' writings: a comparison of handwritten and PCwritten texts

1
Syntactic variables in pupils' writingsa
comparison of hand-writtenand PC-written texts

• Bård Uri Jensen
• University of Bergen / Hedmark University College

2
Contents

• Purpose
• Background theory
• Presentation of text corpus
• Research questions
• Results
• Discussion

3
Purpose / Aim

• Pupils writing in school by hand or on PC
• Does production mode affect syntax ?
• Syntactic variables
• Lexical variables

4
Background theory / previous research

• Word processing
• Russell 1999
• Harrington, Shermis Rollins 2000
• Kellogg Mueller 1993
• Computer-mediated communication
• Baron 1998
• Crystal 2001
• Hård av Segerstad 2002
• Production speed
• Horowitz Berkowitz 1964
• Written and spoken language
• differences resulting from production speed
• Allwood 1998
• Biber 1988
• Halliday 1989

5
Research questions

• How are the following variables affected by
  production mode in pupils writing?
• Lexical density
• Lexical diversity
• Rate of subordination
• Biber 1988, Halliday 1989
• Rate of modal particles
• Rate of certain kinds of topic markers
• Faarlund, Lie Vannebo 1997

6
Research questions

• How are the following variables affected by
  production mode in pupils writing?
• Lexical density
• Rate of subordination
• Biber 1988, Halliday 1989
• Rate of modal particles
• Faarlund, Lie Vannebo 1997

7
Text collection

• 20 pupils in 11th year (16 years old)
• Three hours writing session
• little opportunity for revision / rewriting
• No Internet connection

• Text length

8
Subordination (independent clauses)

• Subjunction count
• At, om, som, fordi, når, så, hvis, hvordan,
• That, whether, which/that, because, when, so
  that, if, how,
• Å ( infinitive)
• To ( infinitive)
• Traces of som and at.
• Han sa ltatgt han skulle komme.
• He said ltthatgt he would come.
• Bilen ltsomgt jeg kjører, er en Toyota.
• The car ltthatgt I drive is a Toyota.
• (Question-type
• Hadde jeg ikke kommet, ville det ikke ha skjedd.
• Had I not come, it would not have happened.)

9
Results Subordination

• Significant differences in subordinations by
  number of (graphic) sentences.

10
Modal particles

• Jo, vel, nok, da, nå, visst
• Jo Known to both sender and receiver.
• Jeg kjører jo Toyota.
• I drive a Toyota, you know.
• Vel Uncertainty and appeals to receivers
  knowledge.
• Jenter leser vel mer bøker.
• Girls read books more, dont they?
• Nok Expresses probability.
• Gutter driver nok mer med data.
• I think boys use their computer more.
• Boys probably use their computer more.

11
Modal particles and text type

• Frequency per 1000 words
• No significant differences related to production
  mode
• Jo, nok are slightly more frequent in PC-texts
• Vel is slightly less frequent in PC-text
• Significant mean differences as function of
  question

12
Modal particles and text length

• Significant positive correlation
• Difference in rate of modal particles with
  production mode
• Total text length produced by pupil
• Pearsons correlation 0.57, slt.01
• Pupils who generally write long texts use more
  modal particles in PC-texts
• Pupils who write long texts
• have good writing skills?
• are motivated?
• utilise speed to produce fluently?
• get carried away?

13
Results Lexical density

• Ratio of lexical words to total words
• Nouns, adjectives and verbs
• Minus function verbs å ha (to have), å være (to
  be)
• Lexical adverbs not included
• Production mode alone shows no influence
• Significant negative correlation between
• Difference in lexical density between production
  modes
• Difference in text length between production
  modes
• Pearsons correlation -.61, slt.01
• No correlation with total text length!

14
Discussion

• Problem of grammatical unit
• Differentiating between different categories of
  pupils
• text length
• text length difference
• Corpus size
• Pupils knowledge of norms?

15
References

• Allwood, Jens (1998). Some Frequency based
  Differences between Spoken and Written Swedish.
  In proceedings from the XVIth Scandinavian
  Conference of Linguistics,
• Department of Linguistics, University of Turku
  Baron, N. S. (1998). Letters by phone or speech
  by other means the linguistics of email.
  Language Communication, 18(2), 133-170.
• Biber, D. (1988). Variation across speech and
  writing. New York Cambridge University Press.
• Crystal, D. (2001). Language and the Internet.
  Cambridge Cambridge University Press.
• Faarlund, J. T., Lie, S., og Vannebo, K. I.
  (1997). Norsk referansegrammatikk. Oslo
  Universitetsforlaget.
• Halliday, M. A. K. (1989). Spoken and written
  language (2nd ed.). Oxford Oxford University
  Press.
• Harrington, S., Shermis, M. D., og Rollins, A. L.
  (2000). The influence of word processing on
  English placement test results. Computers and
  Composition, 17(2), 197-210.
• Horowitz, M. W., og Berkowitz, A. (1964).
  Structural advantage of the mechanism of spoken
  expression as a factor in differences in spoken
  and written expression. Perceptual and motor
  skills, 19, 619-625.
• Hård af Segerstad, Y. (2002). Use and Adaptation
  of Written Language to the Conditions of
  Computer-mediated Communication. Göteborg
  University, Göteborg.
• Kellogg, R. T., og Mueller, S. (1993).
  Performance amplification and process
  restructuring in computer-based writing.
  International Journal of Man-Machine Studies,
  39(1), 33-49.
• Russell, M. (1999). Testing on computers A
  follow-up study comparing performance on computer
  and on paper. Education Policy Analysis Archives,
  7(20).

16
Corpus size

• Difficult to obtain significance
• Some substantial differences / correlations
• Less substantial differences may be significant
  in a larger corpus.

17
Unit of measurement

• Basic principle
• Number of occurances per possible places of use
• Subordination
• Per graphic sentence (i.e. between lt. ! ?gt)
• Should be per independent clause.
• Requires time-consuming manual analysis.
• Modal particles
• Per 1000 words
• Should be per indpendent clause
• Lexical density
• Per total number of words

18
Knowledge of norms

• Long sentences,
• Independent clauses often piled onto each other
• Without conjunctions
• Without full stops
• Without commas, sometimes
• Often seem quite oral in nature
• If pupils dont know the norms, cant be expected
  to strive towards them
• Maybe differences will only show in pupils with
  good writing skills?

19
Categorization of pupils
20
Results Lexical diversity Distribution of word
frequency

• Written
• 10 words 19
• 50 words 38
• 10.000 words 87
• Hand
• 10 words 24
• 50 words 53
• 700 words 91

• Spoken
• 10 words 23
• 50 words 52
• 10.000 words 97 Allwood 1998
• PC
• 10 words 24
• 50 words 53
• 700 words 90

21
Hand PC
22
Hand PC
23
Hand PC
24
Hand PC
25
Hand PC
26
Density in written and spoken language

• Written
• Investment in a rail facility implies a
  long-term commitment.
• Spoken
• If you invest in a rail facility, this implies
  that you are going to be committed for a long
  term.
• Halliday (1989)