Workshop: Week 4

ImportantInstructions

  • Work on Task A as a group, guided by the group navigator, then work on one of the other two tasks. Feel free to work on the remaining task as well if you have time.

  • Each of you should complete the exercises on your laptop, using the RStudio project you created for the course. This includes writing text and code in a Quarto document and executing the code. You should use Quarto documents (rather than scripts) from now on (in the workshops and at home).

NoteTask A: Albanian Voice Onset Time

  • Data docs.

  • Read the coretta2021/alb-vot.csv data in R.

  • Calculate the VOT based on the release and voi_onset columns. (VOT is the time lag between the voicing onset and the consonant release).

  • Create a new column that specifies the voicing of the consonant (voiced or voiceless) based on the value in label. ifelse() or case_when() are useful.

  • Get summary measures of VOT for voiceless and voiced consonants. Can the VOT differentiate between the two voicing categories?

NoteTask B: Albanian vowels

  • Data docs.

  • Read the coretta2021/alb_formants.rds data in R.

  • Create a scatter plot of F1 and F2, with points coloured by vowel. Are the vowels in the places you’d expect them to be?

  • Now create another scatter plot with F1 and vowel duration. You need to calculate vowel duration first from start and end. Can you notice any patterns in the relationship between F1 and vowel duration?

NoteTask C: ‘Everywhere here can say this’

  • Data docs.

  • Read the sluckin2022/everywhere-loc.csv data in R. The data are acceptability judgements on the impersonal use of “everywhere”, like in the sentence “Everywhere here can say this”.

  • The Restrictor column tells you if the sentence has a restrictor, like in “Everywhere on the coast eats fish and chips.” vs no restrictor “Everywhere takes credit cards these days.” The Value column gives you the acceptability rating from 1 to 7.

  • Mutate the Restrictor column so that 0 = “no restrictor” and 1 = “restrictor”.

  • Mutate the Value column so that it is a character column (use as.character()).

  • Create a filled stacked bar chart by Restrictor with fill by Value and panels by Age. Can you spot differences in rating across different age brackets?