Does music soothe the soul? Evaluating the impact of a music education programme in Medellin, Colombia

Abstract

Numerous studies have borne out the effects of cultural and music education on individuals’ well-being, considering music as a mainly systematic practice or skill or as established educational supply. However, few studies assess the impact of music programmes designed to achieve specific goals, where music is considered as a tool for social change. As a case study, we take the Medellin Music School Network (Colombia), whose education programme for music initiation has been running for 23 years. Our aim is to evaluate the economic and social impact generated by participating in this programme. We use a quasi-experimental propensity score matching technique as the evaluation method. Results show that the programme significantly reduces the probability of participants’ becoming involved in conflict, added to which they perceive a better quality of life. Students achieve better academic performance and intensify cultural consumption and participation in artistic activities. Institutional efficacy is reflected through beneficiaries expressing a positive and significant willingness to pay in order to maintain the programme. The work also aims to evidence the usefulness of the methodology for evaluating the impact of cultural policies, particularly in developing areas.

This is a preview of subscription content, access via your institution.

Fig. 1

Source: REMM, www.redmusicamedellin.org

Fig. 2

Source: Own elaboration based on the management reports of the REMM

Fig. 3

Source: Own elaboration

Notes

  1. 1.

    For further details, see www.redmusicamedellin.org.

  2. 2.

    For an assessment reference for this educational project, see the works of Cuesta (2011) and Alemán et al. (2017) who, respectively, apply cost–benefit analysis and experimental techniques (differences in differences) to gauge the social impact of this programme.

  3. 3.

    A summary of the main characteristics and content of the music education programme may be found in “Appendix 1”.

  4. 4.

    According to the programme information, the dropout rate is low and stands at less than 10% per year and is generally due to the academic workload at school or university. One reflection of the level of permanence is that 48% of the present-day teaching staff and musical instrument teachers are graduates of the programme.

  5. 5.

    The 11 integrated ensembles are made up of four symphony orchestras, three choral orchestras and four popular music groups. The presence of popular music groups together with the Colombian music school (Pedregal) enables the REMM to explore autochthonous music, which is reflected in a wide and varied repertoire, ranging from universal classical works to contemporary works composed by school conductors or other musicians in the city, the country and the region.

  6. 6.

    In accordance with the National Department of Statistics in Colombia, the socio-economic stratification of urban areas classifies the residential dwellings that receive public services into groups. It is mainly carried out in order to levy a differential charge, by different strata, for the public services provided to households, thus enabling subsidies to be allocated and contributions to be levied on these areas. This means that those who have the highest economic capacity pay more for public services and contribute so as to enable the lower strata to make their payments, following a progressive fiscal system. The classification in strata in Colombia is as follows: (1) low–low, (2) low, (3) medium–low, (4) medium, (5) medium–high and (6) high.

  7. 7.

    The advantage of PSM is that it only requires information of the outcomes after the programme (cross section data), whereas when applying DD, it is necessary to measure outcomes both before and after the programme (panel data). Moreover, PSM assumes that only observed characteristics can influence participation and outcome, whereas DD can control for unobserved factors. However, for DD to be valid, the comparison group must accurately represent the change in outcomes that would have been experienced by the treatment group in the absence of treatment (i.e. unobserved factors must be time invariant). A more robust method is “matched DD”, which combines PSM and DD (Gertler et al. 2017).

  8. 8.

    We also used a logit specification for the propensity score, and the results did not change significantly. These complementary results are available from the authors upon request. Machine learning methods may also provide an alternative nonparametric approach to propensity score estimation, but they are not suitable for small samples (Pirracchio et al. 2015).

  9. 9.

    We chose an intermediate age to subsequently check (8  years later) the effects on academic improvement and job opportunities.

  10. 10.

    Notable events were also recalled to help jog the interviewees’ memory, such as the IX South American Games or the World Cup held in South Africa that was won by Spain, with the official song sung by the Colombian singer Shakira (Waka Waka). The survey model and the primary results thereof are available upon request from the authors.

  11. 11.

    Commune 1—Popular, Commune 2—Santa Cruz, Commune 3—Manrique, Commune 4—Aranjuez, Commune 5—Castilla, Commune 6—Doce de Octubre, Commune 7—Robledo, Commune 8—Villa Hermosa, Commune 9—Buenos Aires, Commune 10—La Candelaria, Commune 11—Laureles-Estadio, Commune 12—La América, Commune 13—San Javier, Commune 14—Poblado, Commune 15—Guayabal y Commune 16—Belén.

  12. 12.

    All of the estimations performed in this study have been carried out using Stata software.

  13. 13.

    When using the two other matching techniques posited in the research (Caliper and Kernel), the goodness-of-fit tables yield similar results. These complementary results are available from the authors upon request.

  14. 14.

    Later on, in the estimations of the causal effects using the nearest neighbour and Kernel techniques, it can be seen that the sample used indeed corresponds to 176 observations (N = 176), whilst in the case of the nearest neighbour with Caliper (0.01) 13 observations are seen to lie outside the common support region, such that N = 167.

  15. 15.

    It should be pointed out that it was not possible to apply this technique to all the outcome vectors, since the variable must be known both before and after treatment. The early age at which pupils join the REMM in the baseline (on average, when aged 14) means that certain variables (related to cultural consumption, WTP, labour market, and so on) lie outside the individual’s area of decision at the time, although this not the case for the proxy of academic performance (being enrolled in the school year corresponding to their age).

References

  1. Aerts, K., & Schmidt, T. (2008). Two for the price of one? Additionality effects of R&D subsidies: A comparison between Flanders and Germany. Research Policy, 37(5), 806–822.

    Google Scholar 

  2. Alemán, X., Duryea, S., Guerra, N., McEwan, P., Muñoz, R., Stampini, M., et al. (2017). The effects of musical training on child development: A randomized trial of El Sistema in Venezuela. Prevention Science, 18(7), 865–878.

    Google Scholar 

  3. Arias-Pérez, J. A. (2013). Gestión Humana: Actualidad y futuro en la Red de Escuelas de Música de Medellín. Trabajo para optar el título de Especialista en Gestión del Talento Humano y la Productividad. Medellín: Universidad de Antioquia.

  4. Ateca-Amestoy, V. (2009). El capital humano como determinante del consumo cultural. Estudios de Economía Aplicada, 27(1), 87–110.

    Google Scholar 

  5. Ateca-Amestoy, V. (2011). Leisure and subjective well-being. In S. Cameron (Ed.), Handbook on the economics of leisure (pp. 52–76). Cheltenham: Edward Elgar.

    Google Scholar 

  6. Austin, P. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46(3), 399–424.

    Google Scholar 

  7. Backman, M., & Nilson, P. (2018). The role of cultural heritage in attracting skilled individuals. Journal of Cultural Economics, 42(1), 111–138.

    Google Scholar 

  8. Bernabé-Villodre, M. (2017). Music education as a tool to improve socio-emotional and intercultural health within adverse contexts in El Salvador. Procedia—Social and Behavioral Sciences, 237, 499–504.

    Google Scholar 

  9. Berry, C., & Glaeser, E. (2005). The divergence of human capital levels across cities. Papers in Regional Science, 84(3), 407–444.

    Google Scholar 

  10. Bille, T., & Schulze, G. (2006). Culture in urban and regional development. In V. Ginsburgh & D. Throsby (Eds.), Handbook of the economics of art and culture (pp. 1051–1099). Amsterdam: Elsevier.

    Google Scholar 

  11. Blood, A., & Zatorre, R. (2001). Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. PNAS—Proceeding of the National Academy of Sciences, 98(20), 11818–11823.

    Google Scholar 

  12. Boix, R., Hervás, J. L., & De Miguel, B. (2015). Micro-geographies of creative industries clusters in Europe: From hot spots to assemblages. Papers in Regional Science, 94(4), 753–773.

    Google Scholar 

  13. Bourdieu, P. (2000). Las formas del capital. Capital Económico, capital cultural y capital social. In P. Bourdieu (Ed.), Poder, derecho y clases sociales. Barcelona: Desclée.

    Google Scholar 

  14. Broh, B. (2002). Linking extracurricular programming to academic achievement: Who benefits and why? Sociology of Education, 75, 69–95.

    Google Scholar 

  15. Bucci, A., Sacco, P. L., & Segre, G. (2014). Smart endogenous growth: Cultural capital and the creative use of skills. International Journal of Manpower, 35(1/2), 33–55.

    Google Scholar 

  16. Cabane, C., Hille, A., & Lechner, M. (2016). Mozart or Pelé? The effects of adolescents’ participation in music and sports. Labour Economics, 41, 90–103.

    Google Scholar 

  17. Caliendo, M., & Kopeining, S. (2008). Some practical guidance for the implementation of propensity score matching. Journal of Economic Survey, 22(1), 31–72.

    Google Scholar 

  18. Cameron, S. (2016). Past, present and future: Music economics at the crossroads. Journal of Cultural Economics, 40(1), 1–12.

    Google Scholar 

  19. Chen, F., Lui, A., Andrade, H., Valle, C., & Mir, H. (2017). Criteria-referenced formative assessment in the arts. Educational Assessment, Evaluation and Accountability, 29(3), 297–314.

    Google Scholar 

  20. Coll-Martínez, E. J. (2019). Creativity and the city: Testing the attenuation of agglomeration economies in Barcelona. Journal of Cultural Economics. https://doi.org/10.1007/s10824-019-09340-9.

    Article  Google Scholar 

  21. Conejo-Rodríguez, P. A. (2012). El valor formativo de la música para la educación en valores. DEDiCA. Revista de Educação e Humanidades, 2, 263–278.

    Google Scholar 

  22. Corrales-Herrero, H., & Rodríguez-Prado, B. (2016). La incidencia de la formación en la edad adulta en la empleabilidad de los individuos: El caso de España. El trimestre económico, 1(329), 149–183.

    Google Scholar 

  23. Cuesta, J. (2011). Music to my ears: The (many) socioeconomic benefits of music training programmes. Applied Economics Letters, 18, 915–918.

    Google Scholar 

  24. Daykin, N., de Viggiani, N., Pilkington, P., & Moriarty, Y. (2013). Music making for health, well-being and behaviour change in youth justice settings: A systematic review. Health Promotion International, 28(2), 197–210.

    Google Scholar 

  25. Dehejia, R., & Wahba, S. (1999). Causal effects in nonexperimental studies: Reevaluating the evaluation of training programs. Review of Economics and Statistics, 94, 1053–1062.

    Google Scholar 

  26. Dehejia, R., & Wahba, S. (2002). Propensity score-matching methods for nonexperimental causal studies. Review of Economics and Statistics, 84(1), 151–161.

    Google Scholar 

  27. Eerola, P., & Eerola, T. (2014). Extended music education enhances the quality of school life. Music Education Research, 16, 88–104.

    Google Scholar 

  28. Egaña, P., Contreras, D., & Valenzuela, J. (2013). The effect of artistic activities on the development of cognitive and non-cognitive outcomes in Chile: The case of the Curanilahue Youth Orchestra. Santiago de Chile: Centro de Investigación Avanzada en Educación, Universidad de Chile.

    Google Scholar 

  29. Egaña, P., Contreras, D., & Valenzuela, J. (2019). The impact of art-education on human capital: An empirical assessment of a youth orchestra. International Journal of Educational Development, 71, 102105.

    Google Scholar 

  30. Espinal-Monsalve, N. E. (2013). La Disponibilidad a Pagar como una medida de la Legitimidad. El caso de la Biblioteca Pública Piloto para América Latina y los Parques Biblioteca de Medellín, Colombia (Doctoral thesis, Universidad Pablo de Olavide, España).

  31. Falck, O., Fritsch, M., & Heblich, S. (2011). The phantom of the opera: Cultural amenities, human capital, and regional economic growth. Labour Economics, 18(6), 755–766.

    Google Scholar 

  32. Falck, O., Fritsch, M., Heblich, S., & Otto, A. (2018). Music in the air: Estimating the social return to cultural amenities. Journal of Cultural Economics, 42(3), 365–391.

    Google Scholar 

  33. Gardiner, M., Fox, A., Nowles, F., & Jeffrey, D. (1996). Learning improved by arts training. Nature, 381(6580), 284.

    Google Scholar 

  34. Gertler, P. J., Martínez, S., Premand, P., Rawlings, L. B., & Vermeersch, C. (2017). Impact evaluation in practice (2nd ed.). Washington, DC: World Bank.

    Google Scholar 

  35. Gómez-Zapata, J. D., Espinal-Monsalve, N. E., & Herrero-Prieto, L. C. (2018). Economic valuation of museums as public club goods: Why build loyalty in cultural heritage consumption? Journal of Cultural Heritage, 30, 190–198.

    Google Scholar 

  36. Hallam, S. (2010). The power of music: Its impact on the intellectual, social and personal development of children and young people. International Journal of Music Education, 28(3), 269–289.

    Google Scholar 

  37. Hand, C. (2018). Do the arts make you happy? A quantile regression approach. Journal of Cultural Economics, 42(2), 271–286.

    Google Scholar 

  38. Heckman, J., Ichimura, H., Smith, J., & Todd, P. (1998). Characterizing selection bias using experimental data. Econometrica, 66(5), 1017–1098.

    Google Scholar 

  39. Herrero-Prieto, L. C. (2011). La contribución de la cultura y las artes al desarrollo económico regional. Investigaciones Regionales, 19, 177–202.

    Google Scholar 

  40. Herrero-Prieto, L. C., Sanz-Lara, J. A., Bedate-Centeno, A., & del Barrio-Tellado, M. J. (2012). Who pays more for a cultural festival, tourists or locals? A certainty analysis of a contingent valuation application. International Journal of Tourism Research, 14(5), 495–512.

    Google Scholar 

  41. Herrero-Prieto, L. C., Sanz-Lara, J. A., & Devesa-Fernández, M. (2011). Measuring the economic value and social viability of a cultural festival as a tourism prototype. Tourism Economics, 17(3), 639–653.

    Google Scholar 

  42. Hetland, L. (2000). Listening to music enhances spatial-temporal reasoning: Evidence for the ‘Mozart Effect’. Journal of Aesthetic Education, 34, 105–148.

    Google Scholar 

  43. Hille, A., & Schupp, J. (2015). How learning a musical instrument affects the development of skills. Economics of Education Review, 44, 56–82.

    Google Scholar 

  44. Imbens, G., & Wooldridge, J. (2009). Recent developments in the econometrics of program evaluation. Journal of Economic Literature, 47(1), 5–86.

    Google Scholar 

  45. Jalan, J., & Ravallion, M. (2003). Estimating the benefit incidence of an antipoverty program by propensity-score matching. Journal of Business & Economic Statistics, 21(1), 19–30.

    Google Scholar 

  46. Johnson, C. M., & Memmott, J. E. (2006). Examination of relationships between participation in school music programs of differing quality and standardized test results. Journal of Research in Music Education, 54(4), 293–307.

    Google Scholar 

  47. Leuven, E. & Sianesi, B. (2003). psmatch2 version 4.0.12: Stata module to perform full Mahalanobisand propensity score matching, common support graphing, and covariate imbalance testing. Download from http://ideas.repec.org/c/boc/bocode/s432001.html. Accessed January 30, 2016.

  48. Levinson, J. (2014). Values of music. In V. Ginsburgh & D. Throsby (Eds.), Handbook of the economics of art and culture (Vol. 2). Elsevier: North Holland.

    Google Scholar 

  49. Mellander, C., Florida, R., Rentfrow, P., & Potter, J. (2018). The geography of music preferences. Journal of Cultural Economics, 42(4), 593–618.

    Google Scholar 

  50. Michalos, A., & Kahlke, P. (2008). Impact of arts related activities on the perceived quality of life. Social Indicators Research, 89(2), 193–258.

    Google Scholar 

  51. Miksza, P., & Hime, L. (2015). Undergraduate music program alumni’s career path, retrospective institutional satisfaction, and financial status. Arts Education Policy Review, 116(4), 1–13.

    Google Scholar 

  52. Moral-Arce, I. (2014). Elección del método de evaluación cuantitativa de una política pública Buenas prácticas en América Latina y la Unión Europea. Colección Documentos de Trabajo nº 6. Madrid: Programa EUROsociAL.

  53. Ntarangwi, M. (2015). Youth, music, and peace building. In J. Wright (Ed.), International encyclopedia of the social & behavioral sciences (2nd ed, pp. 842–848). Amsterdam: Elsevier.

    Google Scholar 

  54. Pérez-Torres, F. J. (2015). Evaluación de impacto de política pública del Programa Especial Desarraigados: una aplicación cuantitativa. Equidad & Desarrollo, 23, 77–125.

    Google Scholar 

  55. Pettit, B. (1999). Cultural capital and residential mobility: A model of impersistence in place. Poetics, 26(3), 177–199.

    Google Scholar 

  56. Pirracchio, R., Petersen, M. L., & van der Laan, M. (2015). Improving propensity score estimators’ robustness to model misspecification using super learner. American Journal of Epidemiology, 181(2), 108–119.

    Google Scholar 

  57. Rabinowitch, T., Cross, I., & Burnard, P. (2013). Long-term musical group interaction has a positive influence on empathy in children. Psychology of Music, 41(4), 484–498.

    Google Scholar 

  58. Ramos-Vargas, M. (2015). Music as a resource to develop cognition. Procedia—Social and Behavioral Sciences, 174, 2989–2994.

    Google Scholar 

  59. Red de Escuelas de Música de Medellín (REMM) (2010). Retrieved from www.redmusicamedellin.org. Accessed 8 Feb 2019.

  60. Rentfrow, P., & Gosling, S. (2003). The do re mi’s of everyday life: The structure and personality correlates of music preferences. Journal of Personality and Social Psychology, 84(6), 1236–1256.

    Google Scholar 

  61. Rosenbaum, P., & Rubin, D. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55.

    Google Scholar 

  62. Rosenbaum, P., & Rubin, D. (1985). Constructing a control group using multivariate matched sample methods that incorporate the propensity score. American Statistician, 39(1), 33–38.

    Google Scholar 

  63. Schellenberg, E. G. (2011). Examining the association between music lessons and intelligence. British Journal of Psychology, 102(3), 283–302.

    Google Scholar 

  64. Sloboda, J. (2012). La mente musical. La psicología cognitiva de la música. Madrid: Machado Nuevo Aprendizaje.

    Google Scholar 

  65. Smith, J. (2000). A critical survey of empirical methods for evaluating active labor market policies. Swiss Journal for Economics and Statistics, 136(3), 1–22.

    Google Scholar 

  66. Southgate, D., & Roscigno, V. (2009). The impact of music on childhood and adolescent achievement. Social Science Quarterly, 90(1), 4–21.

    Google Scholar 

  67. Swaminathana, S., Schellenberg, E. G., & Khalil, S. (2017). Revisiting the association between music lessons and intelligence: Training effects or music aptitude? Intelligence, 62, 119–124.

    Google Scholar 

  68. Taylor, P., Davies, L., Wells, D., Gilbertson, J. & Tayleur, W. (2015). CASE: A review of the social impacts of culture and sport. Department for Culture, Media and Sport. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/416279/A_review_of_the_Social_Impacts_of_Culture_and_Sport.pdf. Accessed 1 Apr 2019.

  69. Thomas, M. (2016). Music education and its causal impact on student engagement and success. A program evaluation of little kids Rock. Working paper. Retrieved from https://www.arts.gov/sites/default/files/Research-Art-Works-MississippiState.pdf. Accessed 1 Apr 2019.

  70. Tubadji, A. (2012). Culture-based development: Culture as an encompassing economic factor–empirical evidence for Germany. International Journal of Social Economics, 39(9), 690–703.

    Google Scholar 

  71. Wheatley, D., & Bickerton, C. (2017). Subjective well-being and engagement in arts, culture and sport. Journal of Cultural Economics, 41(1), 23–45.

    Google Scholar 

  72. Yang, P. (2015). The impact of music on educational attainment. Journal of Cultural Economics, 39(4), 369–396.

    Google Scholar 

Download references

Acknowledgements

The authors wish to thank the Medellin Music School Network (REMM) for granting access to its databases and information, which allowed the present study to be carried out. We would also like to thank the referee and participants at the 9th Spanish Workshop on Cultural Economics and Management for their comments and discussion on a preliminary version of the paper. We also wish to thank the comments and suggestions made by the anonymous referees of the Journal. The usual disclaimer applies.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Jonathan Daniel Gómez-Zapata.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix 1: Characteristics of the REMM programme

Mission Offering children and youngsters in the city of Medellin an education and a focus through music geared towards instilling an appreciation of life, fostering non-violent attitudes and changing young people through education, healthy coexistence and social appropriation
Character An introduction to music education programme that is extracurricular and independent from the formal education process (school or university)
Publicity General publicity for registration and activities made by the city council and REMM through information channels such as webpage, social networks, television, radio, posters, and so on.
Intake Annual, voluntary and free
Age range between 7 and 24, resident in Medellin and enrolled in a formal education institution. No previous training in music is required
Once registered, students take a music aptitude test to distribute them and form groups of students who have similar levels
Each student must present a family member as a tutor to attend and take part in REMM activities
Organisation 27 schools: 13 bowed string instruments, 13 wind and percussion instruments, and one focusing on Colombian instruments
Four levels: “seedbed”, pre-orchestra or pre-band, orchestra or band, integrated ensembles
11 integrated ensembles: four symphony orchestras, three choral orchestras, and four popular music groups
Wide and varied repertoire based on universal classical works and including contemporary works composed by musicians from the city, the country and the region; this also provides for the possibility of performing autochthonous works
Training Teaching model: collective training programme in an introduction to music and instruments, music language, body expression, instrument technique, choral practice and instrument ensembles; complemented by workshops and psychosocial activities. Students choose their instrument in accordance with the school at which they enrolled
Assessment: system of merit and recognition which mainly involves regular attendance and final presentation at the end of the year of a repertoire chosen together with their teachers. The children and youngsters who most stand out progress through the years until they come to form part of the integrated ensembles
Intensity: group lessons, two sessions per week of approximately 2 h for the basic levels and 8 h a week for the advanced levels
The members of the bands, orchestras and/or ensembles travel and perform together in various local, national and international tours
Costs All of the costs associated with the running and functioning as well as the REMM’s cultural programme are met by the Medellin city council. The activities organised as part of the programme are free for all citizens
Average annual data 5000 students
1000 new students
132 teachers
Groups of 20 students
55,000 h of lessons
4000 own instruments
10% drop out rate
4100 families involved
Budget of two million US dollars
205 direct jobs
500 concerts
45,000 spectators
108 events around the city each year
One international tour
  1. Source: Own elaboration based on REMM

Appendix 2: Measures and definition of variables

Variable Type Measure
Treatment Participation in the REMM Binary = 1 If you participated in the REMM program
Baseline variables (matching variables) Gender Binary = 1 Man
Age Discrete Age of the interviewee in 2010
Educational level
 Secondary (6th to 9th) Binary School year in which the interviewee was in 2010
= 1 Secondary
 High school (10th to 11th) Binary School year in which the interviewee was in 2010
= 1 High school
Family size Discrete Number of people living with the interviewee in 2010
Maximum educational level of the mother
 Elementary school (1st to 5th) Binary = 1 Elementary school
 Secondary (6th to 9th) Binary = 1 Secondary
 High school (10th to 11th) Binary = 1 High school
 Technical Binary = 1 Technical
 Technological Binary = 1 Technological
 Undergraduate (college) Binary = 1 Undergraduate (college)
 Graduate Binary = 1 Graduate
Maximum educational level of the father
 Elementary school (1st to 5th) Binary = 1 Elementary school
 Secondary (6th to 9th) Binary = 1 Secondary
 High school (10th to 11th) Binary = 1 High school
 Technical Binary = 1 Technical
 Technological Binary = 1 Technological
 Undergraduate (college) Binary = 1 Undergraduate (college)
 Graduate Binary = 1 Graduate
Place of residence
 Commune 1—Popular Binary = 1 Popular
 Commune 3—Manrique Binary = 1 Manrique
 Commune 4—Aranjuez Binary = 1 Aranjuez
 Commune 7—Robledo Binary = 1 Robledo
 Commune 8—Villa Hermosa Binary = 1 Villa Hernisa
 Commune 9—Buenos Aires Binary = 1 Buenos Aires
 Commune 10—La Candelaria Binary = 1 La Candelaria
 Commune 11—Laureles-Estadio Binary = 1 Laureles-Estadio
 Commune 12—La América Binary = 1 La América
 Commune 13—San Javier Binary = 1 San Javier
 Commune 15—Guayabal Binary = 1 Guayabal
 Commune 16—Belén Binary = 1 Belén
 Out of Medellin Binary = 1 Out of Medellin
Outcome variables Social efficacy
 Being involved in situations of conflict and/or violence Binary Direct question of the questionnaire
= 1 “You or any of the members of your family group have been involved in a conflict situation (discussion, civil conflict, altercation, violent events, dispute, fight, among others) in the last 5 years”
 Changing place of residence to areas of a higher stratum Binary = 1 If the interviewee changed their place of residence from 2010 to 2018, to an area of greater socio-economic status in the city
 Housing conditions Continuous For that outcome we built an indicator which allowed us to sum up the battery of questions relating to the number of spaces (living room, bedrooms, kitchens, bathroom, among others) and utilities (such as running water, electricity, garbage collection) in a house
Educational and labour efficacy
 Average marks of the studies Continuous Direct question of the questionnaire
“What was the average overall grade you graduated with?”
 Finishing studies in the expected time Binary = 1 If you finish higher studies than the interviewee agreed, in the time established according to your age
 Obtaining study grants Binary Direct question of the questionnaire
= 1 If you receive a scholarship for the development of your studies
 Professional vocation-studying arts Binary = 1 If the interviewee conducted or performs higher studies related to arts
 Entry into the labour market Binary = 1 If the interviewee entered the labour market
 Income above a legal monthly salary Binary = 1 If the interviewee receives a monthly salary higher than the regulated minimum
Cultural consumption
 Doing sports Discrete Direct question of the questionnaire:
“Number of times you performed or attended the following activities during the last year”
 Attending football matches Discrete
 Participating in sports competitions Discrete
 Cinema attendance Discrete
 Theatre attendance Discrete
 Going dancing Discrete
 Museum attendance Discrete
 Book reading Discrete
 Reading newspapers Discrete
 Attending classical music concerts Discrete
 Attending concerts Discrete
 Visiting heritage Discrete
 Videogames Discrete
Institutional efficacy
 Willingness to pay Continuous Direct question of the questionnaire:
“How much would be the voluntary monetary contribution that I would be willing to give once for a year to support the REMM”
 Willingness to support the REMM Binary Direct question of the questionnaire:
= 1 If “Do you think it is important that the community in general be involved with the support of the REMM through a voluntary monetary contribution that allows a greater impact on its activities”
  1. Source: Own elaboration

Appendix 3: Descriptive statistics of the baseline and outcome variables

Sample Complete Treated Controls
N = 180 N = 66 N = 114
Variable Mean SD Min Max Mean SD Min Max Mean SD Min Max
Treatment Participation in the REMM 0.36 0.48 0 1         
Baseline variables (matching variables) Gender 0.5 0.50 0 1 0.46 0.50 0 1 0.51 0.50 0 1
Age 16.43 2.55 13 26 16.21 1.90 14 23 16.56 2.86 13 26
Educational level
 Secondary (6th to 9th) 0.07 0.25 0 1 0.03 0.17 0 1 0.09 0.29 0 1
 High school (10th to 11th) 0.12 0.33 0 1 0.12 0.32 0 1 0.13 0.33 0 1
Family size 4.40 1.41 1 11 4.60 1.49 2 10 4.28 1.35 1 11
Maximum educational level of the mother
 Elementary school (1st to 5th) 0.13 0.34 0 1 0.13 0.34 0 1 0.14 0.34 0 1
 Secondary (6th to 9th) 0.12 0.32 0 1 0.15 0.36 0 1 0.10 0.30 0 1
 High school (10th to 11th) 0.27 0.44 0 1 0.33 0.47 0 1 0.24 0.43 0 1
 Technical 0.16 0.36 0 1 0.15 0.36 0 1 0.16 0.37 0 1
 Technological 0.08 0.28 0 1 0.03 0.17 0 1 0.12 0.32 0 1
 Undergraduate (college) 0.14 0.35 0 1 0.15 0.36 0 1 0.14 0.34 0 1
 Graduate 0.05 0.21 0 1 0.03 0.17 0 1 0.06 0.24 0 1
Maximum educational level of the father
 Elementary school (1st to 5th) 0.20 0.40 0 1 0.28 0.45 0 1 0.14 0.35 0 1
 Secondary (6th to 9th) 0.09 0.29 0 1 0.04 0.20 0 1 0.12 0.32 0 1
 High school (10th to 11th) 0.27 0.44 0 1 0.30 0.46 0 1 0.25 0.43 0 1
 Technical 0.10 0.30 0 1 0.10 0.31 0 1 0.10 0.30 0 1
 Technological 0.07 0.25 0 1 0.03 0.17 0 1 0.09 0.29 0 1
 Undergraduate (college) 0.17 0.38 0 1 0.16 0.37 0 1 0.18 0.38 0 1
 Graduate 0.03 0.19 0 1 0.03 0.17 0 1 0.04 0.20 0 1
Place of residence
 Commune 1—Popular 0.05 0.21 0 1 0.06 0.24 0 1 0.04 0.20 0 1
 Commune 3—Manrique 0.07 0.26 0 1 0.09 0.28 0 1 0.07 0.25 0 1
 Commune 4—Aranjuez 0.04 0.20 0 1 0.07 0.26 0 1 0.02 0.16 0 1
 Commune 7—Robledo 0.06 0.24 0 1 0.07 0.26 0 1 0.05 0.22 0 1
 Commune 8—Villa Hermosa 0.03 0.18 0 1 0.06 0.24 0 1 0.17 0.13 0 1
 Commune 9—Buenos Aires 0.05 0.22 0 1 0.04 0.20 0 1 0.06 0.24 0 1
 Commune 10—La Candelaria 0.05 0.22 0 1 0.04 0.20 0 1 0.06 0.24 0 1
 Commune 11—Laureles-Estadio 0.05 0.21 0 1 0.06 0.24 0 1 0.04 0.20 0 1
 Commune 12—La América 0.04 0.20 0 1 0.03 0.17 0 1 0.05 0.22 0 1
 Commune 13—San Javier 0.03 0.18 0 1 0.04 0.20 0 1 0.02 0.16 0 1
 Commune 15—Guayabal 0.06 0.25 0 1 0.13 0.34 0 1 0.02 0.16 0 1
 Commune 16—Belén 0.04 0.20 0 1 0.03 0.17 0 1 0.05 0.22 0 1
 Out of Medellin 0.27 0.44 0 1 0.18 0.38 0 1 0.32 0.47 0 1
Outcome variables Social efficacy
 Being involved in situations of conflict and/or violence 0.21 0.41 0 1 0.15 0.36 0 1 0.25 0.43 0 1
 Changing place of residence to areas of a higher stratum 0.10 0.30 0 1 0.12 0.32 0 1 0.08 0.28 0 1
 Housing conditions 1.07 0.12 0.69 1.57 1.05 0.13 0.80 1.57 1.07 0.12 0.69 1.38
Educational and labour efficacy
 Average marks of the studies 4.15 0.27 3.5 4.8 4.2 0.30 3.5 4.8 4.12 0.25 3.5 4.8
 Finishing studies in the expected time 0.62 0.48 0 1 0.77 0.42 0 1 0.53 0.50 0 1
 Obtaining study grants 0.18 0.39 0 1 0.25 0.44 0 1 0.14 0.35 0 1
 Professional vocation-studying arts 0.07 0.26 0 1 0.15 0.36 0 1 0.03 0.18 0 1
 Entry into the labour market 0.64 0.48 0 1 0.68 0.46 0 1 0.62 0.48 0 1
 Income above a legal monthly salary 0.35 0.48 0 1 0.39 0.49 0 1 0.33 0.47 0 1
Cultural consumption
 Doing sports 0.41 0.49 0 1 0.33 0.47 0 1 0.45 0.50 0 1
 Attending football matches 4.70 11.24 0 92 2.22 4.94 0 25 6.14 13.43 0 92
 Participating in sports competitions 1.58 3.94 0 30 1.37 2.69 0 16 1.71 4.51 0 30
 Cinema attendance 8.75 10.88 0 100 10.48 13.96 0 100 7.75 8.52 0 50
 Theatre attendance 2.53 5.06 0 40 3.80 6.93 0 40 1.79 3.39 0 20
 Going dancing 1.58 5.86 0 52 2.13 5.15 0 30 1.26 6.23 0 52
 Museum attendance 2.20 5.69 0 50 2.21 2.70 0 10 2.20 6.86 0 50
 Book reading 33.77 75.50 0 365 46.03 88.37 0 365 26.67 66.32 0 365
 Reading newspapers 131.27 149.80 0 365 162.22 153.26 0 365 113.35 145.44 0 365
 Attending classical music concerts 2.20 8.15 0 100 5.31 12.81 0 100 0.40 1.40 0 12
 Attending concerts 2.16 3.23 0 27 3.31 4.48 0 27 1.49 1.92 0 10
 Visiting heritage 2.98 13.64 0 150 1.54 3.10 0 20 3.81 16.95 0 150
 Videogames 28.32 72.93 0 365 36.87 89.12 0 365 23.37 61.55 0 365
Institutional efficacy
 Willingness to pay 51,811.07 85,081.35 0 700 82,749.62 126,179.6 0 700 33,899.28 37,760.02 0 200
 Willingness to support the REMM 0.78 0.41 0 1 0.83 0.37 0 1 0.75 0.43 0 1
  1. Source: Own elaboration

Appendix 4: Diagnostic statistics for one-to-one nearest matching procedure (NN(1))

Variable Sample Mean % Reduct
Treated Control % Biasb Bias T testa p value
Gender Unmatched 0.4697 0.5175 − 9.5   − 0.62 0.539
Matched 0.4838 0.4516 6.4 32.6 − 0.36 0.722
Age Unmatched 16.212 16.561 − 14.4   − 0.88 0.378
Matched 16.258 15.766 20.2 − 40.8 1.38 0.171
Educational level
 Secondary Unmatched 0.0303 0.0964 − 27.3   − 1.66 0.099
Matched 0.0322 0.0249 3.3 87.8 0.27 0.788
 High school Unmatched 0.1212 0.1315 − 3.1   − 0.20 0.842
Matched 0.1290 0.1129 4.8 − 55.6 0.27 0.785
Family size Unmatched 4.6061 4.2895 22.2   1.45 0.148
Matched 4.5968 4.5806 1.1 94.9 0.06 0.953
Maximum educational level of the mother
 Elementary school Unmatched 0.1363 0.1403 − 1.1   − 0.07 0.941
Matched 0.1451 0.0806 18.6 − 1518.1 1.13 0.260
 Secondary Unmatched 0.1515 0.1052 13.8   0.91 0.364
Matched 0.1290 0.1532 − 7.2 47.7 − 0.38 0.702
 High school Unmatched 0.3333 0.2456 19.3   1.26 0.208
Matched 0.3387 0.3951 − 12.4 35.6 − 0.65 0.518
 Technical Unmatched 0.1515 0.1666 − 4.1   − 0.27 0.791
Matched 0.1451 0.1129 8.8 − 112.9 0.53 0.596
 Technological Unmatched 0.0303 0.1228 − 35.2   − 2.12 0.036
Matched 0.0322 0.0564 − 9.2 73.8 − 0.65 0.517
 Undergraduate Unmatched 0.1515 0.1403 3.1   0.20 0.838
Matched 0.1612 0.1612 0.0 100.0 0.00 1.000
 Graduate Unmatched 0.0303 0.0614 − 14.8   − 0.92 0.359
Matched 0.0322 0.0249 3.8 74.1 0.27 0.788
Maximum educational level of father
 Elementary school Unmatched 0.2878 0.1912 33.8   2.26 0.025
Matched 0.2741 0.1935 19.7 41.9 1.06 0.293
 Secondary Unmatched 0.0454 0.1228 − 28.0   − 1.71 0.088
Matched 0.0483 0.0806 − 11.7 58.3 − 0.73 0.469
 High school Unmatched 0.3030 0.2543 10.8   0.70 0.483
Matched 0.2903 0.3306 − 9.0 17.1 − 0.48 0.631
 Technical Unmatched 0.1060 0.1052 0.3   0.02 0.987
Matched 0.1129 0.1693 − 18.3 − 6979.0 − 0.90 0.371
 Technological Unmatched 0.0303 0.0964 − 27.3   − 1.66 0.099
Matched 0.0322 0.0161 6.6 75.6 0.58 0.563
 Undergraduate Unmatched 0.1666 0.1842 − 4.6   − 0.30 0.768
Matched 0.1774 0.0967 21.1 − 359.7 1.30 0.195
 Graduate Unmatched 0.0303 0.0438 − 7.1   − 0.45 0.652
Matched 0.3226 0.4839 − 8.5 − 19.0 − 0.45 0.651
Place of residence
 Commune 1—Popular Unmatched 0.0606 0.0438 7.5   0.49 0.622
Matched 0.0645 0.0887 − 10.8 − 44.5 − 0.50 0.616
 Commune 3—Manrique Unmatched 0.09091 0.0701 7.6   0.50 0.619
Matched 0.0967 0.0645 11.8 − 55.6 0.66 0.513
 Commune 4—Aranjuez Unmatched 0.0757 0.0263 22.5   1.55 0.122
Matched 0.0806 0.0645 7.3 67.4 0.34 0.732
 Commune 7—Robledo Unmatched 0.0757 0.0526 9.4   0.62 0.535
Matched 0.0806 0.1048 − 9.8 − 4.6 − 0.46 0.646
 Commune 8—Villa Hermosa Unmatched 0.0606 0.0175 22.2   1.55 0.122
Matched 0.0483 0.0645 − 8.3 62.5 − 0.39 0.700
 Commune 9—Buenos Aires Unmatched 0.0454 0.0614 − 7.1   − 0.45 0.655
Matched 0.0483 0.0564 − 3.6 49.4 − 0.20 0.842
 Commune 10—La Candelaria Unmatched 0.0454 0.0614 − 7.1   − 0.45 0.655
Matched 0.0483 0.0564 − 3.6 49.4 − 0.20 0.842
 Commune 11—Laureles-Estadio Unmatched 0.0606 0.0438 7.5   0.49 0.622
Matched 0.6452 0.0725 − 3.6 51.8 − 0.18 0.860
 Commune 12—La América Unmatched 0.0303 0.0526 − 11.2   − 0.70 0.486
Matched 0.0322 0.0322 0.0 100.0 − 0.00 1.000
 Commune 13—San Javier Unmatched 0.0454 0.0263 10.2   0.69 0.493
Matched 0.0483 0.0241 12.9 − 26.4 0.72 0.475
 Commune 15—Guayabal Unmatched 0.1363 0.0263 40.8   2.90 0.004
Matched 0.0967 0.0806 6.0 85.3 0.31 0.755
 Commune 16—Belén Unmatched 0.0303 0.0563 − 11.2   − 0.70 0.486
Matched 0.0322 0.1613 8.1 27.8 0.58 0.563
 Out of Medellin Unmatched 0.1818 0.3245 − 33.1   − 2.09 0.038
Matched 0.1935 0.2338 − 9.3 71.8 − 0.54 0.588
  1. Source: Own elaboration
  2. at test for equal means between treated and control samples. p values < 0.10 indicate significant mean differences at 10%. p values > 0.10 indicate non-significant differences
  3. bStandardised percentage bias

Appendix 5: Aggregated diagnostic statistics for matching procedures

  Pseudo R2a LR testb p > χ2b Mean biasc Median biasc Out of common supportd
Unmatched sample 0.197 46.70 0.045 14.9 11.0  
Matched sample
 NN(1) 0.075 12.88 0.999 8.9 8.4 4
 NN(1) with Caliper 0.107 15.66 0.993 10.2 7.6 13
 Kernel 0.042 7.30 1.000 6.8 5.9 4
  1. Source: Own elaboration
  2. aPseudo-R2 of propensity score model
  3. bLikelihood ratio test of insignificance of all regressors and p value
  4. cMean and median bias are summary indicators of the distribution of the abs(bias)
  5. dTreated observation out of common support

Appendix 6: Average treatment effects (matched differences in differences)

Finishing studies within the expected time Treated Controls ATT T stat
NN(1) N = 176 − 0.04838 − 0.3225 0.2741 2.21**
NN(1) with Caliper N = 167 − 0.0566 − 0.3396 0.2830 2.15**
Kernel N = 176 − 0.0483 − 0.2845 0.2361 2.49**
  1. Source: Own elaboration
  2. *p value < 0.1; **p value < 0.05; ***p value < 0.01

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Gómez-Zapata, J.D., Herrero-Prieto, L.C. & Rodríguez-Prado, B. Does music soothe the soul? Evaluating the impact of a music education programme in Medellin, Colombia. J Cult Econ 45, 63–104 (2021). https://doi.org/10.1007/s10824-020-09387-z

Download citation

Keywords

  • Cultural economics
  • Music training
  • Cultural policy evaluation
  • Propensity score matching
  • Efficacy

JEL Classification

  • Z11
  • Z18
  • H53
  • O12
  • C13