This readme file was generated on 2024-01-10 by Thomas Anthony GRAVES 


GENERAL INFORMATION

Title of Dataset: A genre-specific measure of subjective feeling in music listening: Constructing the Durham Emotion in Qawwālī Scale (DEQS) [dataset]


Author/Principal Investigator Information
Name: Thomas Anthony GRAVES 
ORCID: 0000-0002-5398-3819
Institution: Durham University
Address: Department of Music, Durham University, Palace Green, Durham, UK, DH1 3RL
Email: thomas.a.graves@durham.ac.uk, thomas.graves@glasgow.ac.uk, thomasgraves11@gmail.com

Author/Associate or Co-investigator Information
Name: Tuomas EEROLA
ORCID: 0000-0002-2896-929X
Institution: Durham University
Address: Department of Music, Durham University, Palace Green, Durham, UK, DH1 3RL
Email: tuomas.eerola@durham.ac.uk

Author/Alternate Contact Information
Name: Martin CLAYTON
ORCID: 0000-0002-9670-5077
Institution: Durham University
Address: Department of Music, Durham University, Palace Green, Durham, UK, DH1 3RL
Email: martin.clayton@durham.ac.uk

Author/Alternate Contact Information
Name: Syed Murshid NIZAMI
ORCID: NA
Institution: NA
Address: Hazrat Nizamuddin Auliya Dargah, Baoli Gate Road, Nizamuddin Basti, Nizamuddin West, New Delhi, Delhi 110013, India
Email: syedmurshidnizami@gmail.com

Author/Alternate Contact Information
Name: Muhammad Usama RAFIQ
ORCID: NA
Institution: NA
Address: NA
Email: Osamarafiqrafiq@gmail.com

Date of data collection: 2021-08-03 to 2022-04-18 

Geographic location of data collection: New Delhi, India

Information about funding sources that supported the collection of the data: A Society for Education and Music Psychology Research (SEMPRE) Gerry Farrell Travelling Scholarship partially supported a field visit during which the corresponding author collected some of the questionnaire responses in this dataset.


SHARING/ACCESS INFORMATION

Licenses/restrictions placed on the data: CC BY NC

Links to publications that cite or use the data: NA

Links to other publicly accessible locations of the data: NA

Links/relationships to ancillary data sets: NA

Was data derived from another source?
If yes, list source(s): No

Recommended citation for this dataset: Graves, T.A., Eerola, T., Clayton, M., Nizami, S.M., Rafiq, M.U. (2024): A genre-specific measure of subjective feeling in music listening: Constructing the Durham Emotion in Qawwālī Scale (DEQS) [dataset]. Durham University. (dataset)


DATA & FILE OVERVIEW

File List: 

There are two folders, one for each questionnaire study:

Questionnaire_1
Questionnaire_2

The files within these folders are entitled as below:

	Folder			File name	   					  	 Description
	----------------------------------------------------------------------------------------------------------------
	Questionnaire_1		Thomas_Graves_Qawwali_Emotion_Concepts_Study_1.csv		 Raw data from Qualtrics 
											   	 questionnaire one - 													 Likert scale ratings of 												 relevance of emotion 													 words for qawwali.

	Questionnaire_1		Discrete_emotions_which_emotions_at_qawwali_2.R		   	 RStudio Script for 													 processing 														 questionnaire one.

	Questionnaire_1		Tables_from_study_1.xlsx					 Tables showing 													 descriptive stats of 													 questionnaire one 												  	 ratings in excel 													 spreadsheet.

	Questionnaire_2		Qualtrics_DEQS_study_2_numeric_18_April_2022_no_header_1.csv  	 Raw data from Qualtrics 
											   	 questionnaire two - 													 Likert scales comparing 												 all possible 														 combinations of 20 													 emotion words. 													 Numerical only.

	Questionnaire_2		Qualtrics_DEQS_study_2_data_words_18_April_2022_highlighted.csv  As above, but with full 
												 questions as presented 												 to participants rather 											         than only numbers.

	Questionnaire_2		DEQS_MDS_FA_12.R						 RStudio Script for 													 processing
											   	 questionnaire two.

	Questionnaire_2		convert2matrix.R					   	 RStudio Script for use 												 in above R file for 													 data structuring.



Relationship between files, if important: convert2matrix is necessary for DEQS_MDS_FA_12 to function. The .R files are used to process the .csv files.

Additional related data collected that was not included in the current data package: A later questionnaire was conducted using the DEQS scale outputted from this data as part of Thomas A. Graves' PhD thesis.

Are there multiple versions of the dataset? Yes - both versions of the questionnaire 2 dataset are included here. There is an earlier version of the dataset 1 data downloaded from Qualtrics including IP addresses and location data.
If yes, name of file(s) that was updated: Qualtrics_DEQS_study_2_numeric_18_April_2022_no_header_1.csv, Qualtrics_DEQS_study_2_data_words_18_April_2022_highlighted.csv, Thomas_Graves_Qawwali_Emotion_Concepts_Study_1.csv
Why was the file updated? To transform all word answers to numerical for analysis, and remove IP addresses and location data to ensure full anonymity of participants. The questionnaire 1 data was also edited to remove IP addresses and location data.
When was the file updated? Two different versions of the file were downloaded from qualtrics - the numerical and non-numerical during analysis (2022). Both files were edited to remove location data (December 2023)


METHODOLOGICAL INFORMATION

Description of methods used for collection/generation of data: 


Study 1: The questionnaire was run online through Qualtrics. 31 participants were recruited at Hazrat Nizamuddin Dargah by Syed Murshid Nizami, where they completed the questionnaire on his smartphone. All participants were regular attendees of qawwali at the shrine. Participants were presented with 75 emotion terms including Sufi terms and Urdu translations of sixty-six adjectives  shown to music festival attendees in the third study of Zentner and colleagues’ GEMS article (2008: 503). Participants were asked to rate how often they feel each of these terms while listening to qawwali on a likert scale of 1-5.

Study 2: In study two, the lower number of terms outputted by study one (totalling emotion 20 terms) were used. 32 participants were recruited at Hazrat Nizamuddin Dargah by Syed Murshid Nizami and Thomas A. Graves, where they completed the questionnaire on smartphones. Participants were presented with all possible combinations of two of the twenty emotion terms and asked to rate how similar those two terms feel on a scale of 1-5 (a total of 191 questions). Due to the large number of questions, three attention checks were included, in which participants were asked to select a certain number. Three participants failed at least one attention check so these answers were excluded from analysis. Another participant answered “2” for most questions, so that response was also excluded, leaving a total N of 28 for analysis.

Methods for processing the data: 

Study 1: Means and Standard deviations for each term were calculated, and terms scoring a mean average above 4 were retained for use in study 2. Several theoretically important terms for qawwali were also retained for use in study 2.

Study 2: First, the comparisons between emotion terms were organised into a dissimilarity matrix for each participant. Then 2D multidimensional scaling was conducted for each participant individually. After this, an aggregated 2D multidimensional scaling was conducted to combine all of the participants' results. Lines were drawn between the most closely related results on the 2D solution based on reading of the data. Next, 3D SMACOF Multidimensional scaling was conducted. Finally, exploratory factor analysis was conducted on the data, determining how many factors could explain the data, and which emotion terms were loaded onto which factor.

Instrument- or software-specific information needed to interpret the data: 

Software - RStudio Version 2023.09.1+494 (2023.09.1+494)
	Required packages - dplyr, smacof, readxl, ggplot2, rgl, psych, gdata

Standards and calibration information, if appropriate: NA

Environmental/experimental conditions: Online delivery of questionnaire.

Describe any quality-assurance procedures performed on the data: Prior to data collection, both questionnaires were checked by Muhammad Usama Rafiq for translation errors in the Urdu. Syed Murshid Nizami also checked both questionnaires for any questions which did not make sense in the context of qawwali at Hazrat Nizamuddin Auliya Dargah, where he is a custodian. In the first questionnaire, after the first participants completed it, Syed Murshid Nizami communicated to Thomas A. Graves that the participant had commented that some of the words were "the same" therefore, the phrases ‘maiṅ kaifiyat tārī huṅ’ (‘I am in an (altered) state (of consciousness)’), and ‘halki kaifiyat’ (light trance), were removed, leaving only "kaifiyat" to describe these terms. Several changes were made to the translations prior to beginning data collection. In study 2, three attention checks were included, and four responses excluded from analysis on the basis of either failing attention checks or repeating the same answer to all questions. For both questionnaires, participants were given the opportunity to add important terms which had been missed out in a free text box at the end of the questionnaire. While some used this to describe their emotions during qawwali, there were no comments requiring additional terms to be added.

People involved with sample collection, processing, analysis and/or submission: 
	Data collection - Syed Murshid Nizami, Thomas A. Graves
	Data processing and analysis - Thomas A. Graves, Tuomas Eerola




DATA-SPECIFIC INFORMATION FOR: Thomas_Graves_Qawwali_Emotion_Concepts_Study_1.csv

Number of variables: 82 (this is the number of questions asked, including demographics but not including consent questions or automated identifier columns)

Number of cases/rows: 31 (not including name of column, second name of column, and ImportID rows)

Variable List: Study is exploratory and thus does not have clear dependent and independent variables. Demographic data collected are: regular attendance at the shrine?, age, religion, gender, devotee of the saint?. Emotion data collected were presented as a table of 75 terms. A full list of 75 Urdu/Arabic/Farsi terms and their English translations can be found in tab one of the spreadsheet "Tables_from_study_1.xlsx".

Missing data codes: <list code/symbol and definition>

Specialized formats or other abbreviations used: Questions and answers presented in both Urdu and English, separated by a forward slash "/". Likert scale answers regarding each emotion term are given as Arabic numerals followed by a romanised Urdu indication of the meaning of each number in parentheses.



DATA-SPECIFIC INFORMATION FOR: Discrete_emotions_which_emotions_at_qawwali_2.R	

Number of variables: NA - this is script for data processing

Number of cases/rows: NA - this is script for data processing

Variable List: NA - this is script for data processing

Missing data codes: NA - this is script for data processing

Specialized formats or other abbreviations used: NA - this is script for data processing



DATA-SPECIFIC INFORMATION FOR: Tables_from_study_1.xlsx

Number of variables: NA - this is an excel spreadsheet organising the output of the R script for study 1. The first sheet shows means and standard deviations for each of 75 emotion terms. The second sheet shows 14 emotion terms scoring means above 4. The third sheet includes these 14 words with 6 theoretically important terms, making the total of 20 terms to be included in study 2.

Number of cases/rows: Sheet 1: 75

Variable List: NA

Missing data codes: NA

Specialized formats or other abbreviations used: NA



DATA-SPECIFIC INFORMATION FOR: Qualtrics_DEQS_study_2_numeric_18_April_2022_no_header_1.csv

Number of variables: 200 (this is the number of questions asked, including demographics but not including consent questions or automated identifier columns)

Number of cases/rows: 32

Variable List: Study is exploratory and thus does not have clear dependent and independent variables. Demographic data collected are: regular attendance at the shrine?, age, religion, gender, devotee of the saint?. There are 191 questions asking participants to compare how two emotion words feel. Three attention checks are spread evenly among these 191 questions.The 20 emotion words are those found in the table on the third sheet of "Tables_from_study_1.xlsx".

Missing data codes: NA

Specialized formats or other abbreviations used: All data are presented numerically only for ease of processing. For full questions and answers as presented to participants see "Qualtrics_DEQS_study_2_data_words_18_April_2022_highlighted".



DATA-SPECIFIC INFORMATION FOR: Qualtrics_DEQS_study_2_data_words_18_April_2022_highlighted

Number of variables: 200 (this is the number of questions asked, including demographics but not including consent questions or automated identifier columns)

Number of cases/rows: 32

Variable List: Study is exploratory and thus does not have clear dependent and independent variables. Demographic data collected are: regular attendance at the shrine?, age, religion, gender, devotee of the saint?. There are 191 questions asking participants to compare how two emotion words feel. Three attention checks are spread evenly among these 191 questions.The 20 emotion words are those found in the table on the third sheet of "Tables_from_study_1.xlsx". This is the same data table as "Qualtrics_DEQS_study_2_numeric_18_April_2022_no_header_1.csv" but with the full questions as presented to participants.

Missing data codes: NA

Specialized formats or other abbreviations used: Questions and answers presented in both Urdu and English, separated by a forward slash "/". Likert scale answers regarding each comparison between emotion terms are given as Arabic numerals followed by a romanised Urdu indication of the meaning of each number in parentheses.



DATA-SPECIFIC INFORMATION FOR: DEQS_MDS_FA_12.R

Number of variables: NA - this is script for data processing

Number of cases/rows: NA - this is script for data processing

Variable List: NA - this is script for data processing

Missing data codes: NA - this is script for data processing

Specialized formats or other abbreviations used: NA - this is script for data processing



DATA-SPECIFIC INFORMATION FOR: convert2matrix.R

Number of variables: NA - this is script for data processing

Number of cases/rows: NA - this is script for data processing

Variable List: NA - this is script for data processing

Missing data codes: NA - this is script for data processing

Specialized formats or other abbreviations used: NA - this is script for data processing