I had to create a word list for a speech intelligibility assessment I am completing. A previous relatively large scale study has analysed the phoneme distribution in % of the language. I need to compare the phoneme distribution (% for each sound) of my wordlist to the phoneme distribution of the large study. Which statistical test should I use to analyse if they are similar to each other and hence ascertain that my list approximates the distribution? Thanks
Hello Pasquale,
If you'd like to use the prior study results as the benchmark, then a chi-squared goodness of fit test would help you determine whether your observed frequencies matched those expected for each phoneme from the benchmark data. In SPSS, the chi-square G-O-F test appears under the "Nonparametric" option of the Analysis menu (if you can not find it straight away, go to "Legacy dialogs" under "Nonparametric").
Example: If the benchmark data showed the phoneme "/F/" (as in fan) appeared 6% of the time, and your data set shows that it appeared 190 times out of a word list yielding 5000 word sounds, then the observed frequency (190) is somewhat mismatched to the expected frequency (6% of 5000 = 300).
Good luck with your work!
Dear David,
Thank you for your reply and feedback. Can I just compare the percentages with the chi-squared goodness of fit? Or I also need to make use of the respective phoneme frequencies for each list?
Thank you
Pasquale,
Frequencies: The basic formula is chi-square = sum[(fo - fe)^2 / fe]
where: fo = observed frequency for each phoneme
fe = expected frequency (as explained above, in first message)
and values are summed across all phonemes
Hi Pasquale,
I faced a similar situation - i wanted to compare the phoneme frequency of patient (people with aphasia) productions against normal phoneme frequencies. If you want to look at the overall distributions then you could use a kolmorogov-smirnov 2 test which should tell you whether your phoneme distributions differ generally (i.e. whether the zipfian type distribution we see typically in English phoneme frequency is adhered to in your word list). This will not tell you whether specific phonemes are over represented though. You could also use a correlation analysis. If you wanted to analyse each phoneme separately then maybe a chi- square would be helpful.
Hope this is useful,
Emma.
- Badges
- Science topic
- Similar topics
- Political Science
- International History and Politics
- Globalization