All_formants_July_2018.xlsx --------------------------- This file is the master file containing the aggregated data for 120 individual participants (denoted P then participant number). There are 16,390 rows in the spreadsheet. The data are comprised of metadata about the participant and the group to which s/he belongs, the software settings, and characteristics of the recording from which each observation was drawn. The key data are vowel formant frequency measurements, in Hertz, for the peripheral monophthongs denoted by the lexical set keywords FLEECE, TRAP, START/PALM, BATH, DRESS, and NORTH/FORCE (after Wells 1982). In the varieties of English represented in the TUULS corpus, SQUARE is often a monophthong (represented here by [e:]) rather than an ingliding diphthong. We have also included formant measurements for the relevant vowels of the commA and lettER sets, plus tokens of schwa which are not exemplars of commA or lettER. Participants were asked to read a short text passage ('Fern's Star Turn') and an approximately 200-item wordlist presented in a word-by-word format on a laptop computer screen in the form of a PowerPoint presentation. They were recorded using professional-quality solid-state audio recorders and head-mounted microphones, in sound-treated laboratory conditions (commercial-grade recording facilities). Spreadsheet contents, by column heading (letter in brackets, e.g. (A)) ====================================================================== (A) Speaker -------- P = participant + number, e.g. P12. (B) vowel TERM vs NURSE vs BIRD ------------------------------- This column is provided as a means of distinguishing between tokens of the NURSE lexical set. Prior to the merger of the short pre-rhotic vowels (the NURSE merger), the words 'bird', 'berth' and 'burn' had distinct vowels, as can still be found in some Scottish and Irish varieties. So as to investigate whether there were any remnants of the pre-merger contrasts in the three varieties represented in the TUULS corpus, we subdivided our set of NURSE items into three groups: TERM, NURSE, and BIRD. We denote these in column (B) as follows: TERM = er NURSE = ur BIRD = ir See also the description of the contents of columns (Q) and (R), below. (C) context ----------- The word in which the target vowel is found, shown in capital letters. In polysyllabic words, the target vowel is put in brackets, for example C(A)NNOT, P(EO)PLE. See also the description of the contents of column (S), below. (D) check --------- Denotes items that were checked for accuracy. Corrections are shown in the spreadsheet accessed using the second tab marked 'corrections'. (E) group --------- The speaker group the individual participant belongs to. Format is Locality/Sex/Age/Mobility: M = Middlesbrough N = Newcastle S = Sunderland M = male F = female O = older Y = young M = mobile N = non-mobile e.g. MMOM = Middlesbrough Male Older Mobile SFYN = Sunderland Female Young Non-mobile (F) type -------- Denotes one of two reading task types: WL = word list RP = reading passage (G) F1 ------ Averaged value in Hz of the first formant over the selected portion. (H) F2 ------ Averaged value in Hz of the second formant over the selected portion. (I) F3 ------ Averaged value in Hz of the third formant over the selected portion. (J) f3-f2 --------- Value of F2 subtracted from value of F3 (Hz) [currently blank] (K) f3-f1 --------- Value of F1 subtracted from value of F3 (Hz) [currently blank] (L) f2-f1 --------- Value of F1 subtracted from value of F2 (Hz) [currently blank] (M) From -------- Timestamp marking beginning of selected portion over which formant values were measured (in seconds from the beginning of the audio file). (N) To ------ Timestamp marking end of selected portion over which formant values were measured (in seconds from the beginning of the audio file). (O) vowel/frame --------------- ASCII conventions representing IPA characters. The correspondences between these sound categories and Wells' lexical set keywords are set out below. a = TRAP a: = START/PALM a2 = BATH c = schwa (other than lettER/commA) c: = NURSE e = DRESS e: = SQUARE i: = FLEECE K = commA or = NORTH/FORCE v = lettER After Wells, J.C. (1982) Accents of English, 3 vols. Cambridge: CUP. (P) IPA ------- Correspondences between the contents of column (O) and symbols of the International Phonetic Alphabet, and lexical set keyword labels for English vowel phonemes devised by Wells (1982). (Q) vowel TERM vs NURSE/BIRD ---------------------------- This column relates exclusively to tokens of the NURSE lexical set. See the above description of the contents of column (B), and that for column (R), below. Prior to the merger of the short pre-rhotic vowels (the NURSE merger), the words 'bird', 'berth' and 'burn' had distinct vowels, as can still be found in some Scottish and Irish varieties. So as to investigate whether there were any remnants of the pre-merger contrasts in the three varieties represented in the TUULS corpus, we subdivided our set of NURSE items into three groups: TERM, NURSE, and BIRD. As a first pass, we distinguished TERM items (e.g. 'person', 'served', 'Fern') from all other NURSE-class items (including BIRD-class items such as 'fir' and 'first'). We denote these as follows: TERM = er NURSE/BIRD = c: (R) vowel TERM vs NURSE vs BIRD ------------------------------- This column relates exclusively to tokens of the NURSE lexical set. See the above descriptions of the contents of columns (B) and (Q). Prior to the merger of the short pre-rhotic vowels (the NURSE merger), the words 'bird', 'berth' and 'burn' had distinct vowels, as can still be found in some Scottish and Irish varieties. So as to investigate whether there were any remnants of the pre-merger contrasts in the three varieties represented in the TUULS corpus, we subdivided our set of NURSE items into three groups: TERM, NURSE, and BIRD. We denote these as follows: TERM = er NURSE = c: BIRD = ir Note that the contents of column (R) are identical to those of column (B). (S) context ----------- The contents of this column are identical to those of column (C). (T) Max_Formant --------------- Upper threshold set in Praat: Formant > Maximum formant (Hz): Typical value 5000Hz (U) No_of_formants ------------------ Number of formants set in Praat: Formant > Number of formants: Typical value 4 (V) Window length ----------------- Window length set in Praat: Formant > Window length (s): Default value 0.025 (W) Dynamic range ----------------- Dynamic range set in Praat: Formant > Dynamic range (dB): Default value 35.0 (X) Method ---------- Method set in Praat Default value Burg (Y) Pre-emph ------------ Pre-emphasis set in Praat Default value 50 If you have any questions about the contents of the file please contact Dom Watt (dominic.watt@york.ac.uk) or Carmen Llamas (carmen.llamas@york.ac.uk). 1st October 2019