Alternative analyses

In the paper we utilise the division of WEIRD/Non-WEIRD countries suggested by Krys et al. 2024.

knitr::kable(table(d$CountryDataCollected_WEOG)/sum(table(d$CountryDataCollected_WEOG)),digits = 2)
Var1 Freq
Non-WEOG 0.23
WEOG 0.77

We considered two possible alternatives earlier, one based on the UN definition of the Western European and Others Group (WEOG) and one based on cultural distances (Mutkukhrisna et al., 2020). In short, using these divisions to allocate WEIRD and non-WEIRD does not materially change our analyses or results. Here are the main summaries conducted with these two alternative country divisions.

Descriptives for alternative WEIRD formulations: WEOG countries

Western European and Others Group (WEOG) defined by UN, https://en.wikipedia.org/wiki/Western_European_and_Others_Group.

source('scripts/alternative_WEIRD_indices.R')
d1 <- alternative_WEIRD_indices(d,'WEOG'); d<-d1$d; DF<-d1$DF

Using WEIRD country index: WEOG

knitr::kable(table(d$CountryDataCollected_WEOG)/sum(table(d$CountryDataCollected_WEOG)),digits = 2)
Var1 Freq
Non-WEOG 0.16
WEOG 0.84 
source('scripts/table1.R')
Sample Size
sample_country_data_collected_WEOG median lwr.ci upr.ci
WEOG 48 44 52
Non-WEOG 60 49 76

[1] “\(W = 120,207.00\), \(p = .030\)”

Age Mean
wtd.avg CI.LL CI.UL
Non-WEIRD 20.32 17.25 22.71
WEIRD 27.97 25.67 30.32

[1] “t value = 9.5, df = 242.07, p-value = 0”

Age SD:

Age SD
wtd.avg CI.LL CI.UL
Non-WEIRD 4.38 3.25 5.28
WEIRD 7.21 6.23 8.30

[1] “t value = 7.21, df = 189.32, p-value = 0”

Gender balance (country data collected):

Gender balance (Primary Country Data Collected)
wtd.avg CI.LL CI.UL
Non-WEIRD 0.55 0.50 0.58
WEIRD 0.59 0.57 0.61

[1] “t value = 2.6, df = 150.57, p-value = 0.006”

Gender balance (based on first author country):

Gender balance (Primary Author Country)
wtd.avg CI.LL CI.UL
Non-WEIRD 0.56 0.50 0.59
WEIRD 0.58 0.57 0.60

[1] “t value = 2, df = 121.93, p-value = 0.034”

Solely musicians:

Musical expertise across samples
Expertise %
musicians 30.5
musicians; non-musicians 22.0
non-musicians 10.1
Not specified 37.4

[1] “musicians n=371: 30% [27% to 33%]” [1] “musicians n=58: 33% [26% to 40%]”

Solely musicians
var n label prop lwr.ci upr.ci
musicians 371 WEIRD 0.30 0.27 0.33
others 863 WEIRD 0.70 0.67 0.73
musicians 58 Non-WEIRD 0.33 0.26 0.40
others 119 Non-WEIRD 0.67 0.60 0.74

[1] “Chi = 0.53 , p-value = 0.470764617691154”

musicians and non-musicians: [1] “musicians n=654: 53% [50% to 56%]” [1] “musicians n=88: 50% [42% to 57%]”

musicians and non-musicians
var n label prop lwr.ci upr.ci
musicians 654 WEIRD 0.53 0.50 0.56
non-musicians 580 WEIRD 0.47 0.44 0.50
musicians 88 Non-WEIRD 0.50 0.42 0.57
non-musicians 89 Non-WEIRD 0.50 0.43 0.58

[1] “Chi = 0.67 , p-value = 0.407796101949025”

Solely Non-musicians: [1] “non-musicians n=127: 10% [9% to 12%]” [1] “non-musicians n=26: 15% [10% to 20%]”

non-musicians
var n label prop lwr.ci upr.ci
others 1107 WEIRD 0.90 0.88 0.91
non-musicians 127 WEIRD 0.10 0.09 0.12
others 151 Non-WEIRD 0.85 0.81 0.91
non-musicians 26 Non-WEIRD 0.15 0.10 0.20

[1] “Chi = 3.1 , p-value = 0.0954522738630685”

University sample: [1] “university n=506: 41% [38% to 44%]” [1] “university n=73: 41% [34% to 49%]”

University samples
var n label prop lwr.ci upr.ci
others 728 WEIRD 0.59 0.56 0.62
university 506 WEIRD 0.41 0.38 0.44
others 104 Non-WEIRD 0.59 0.51 0.66
university 73 Non-WEIRD 0.41 0.34 0.49

[1] “Chi = 0 , p-value = 1”

Sample unspecified: [1] “unsp n=552: 45% [42% to 48%]” [1] “unsp n=69: 39% [32% to 47%]”

Unspecified samples
var n label prop lwr.ci upr.ci
others 682 WEIRD 0.55 0.52 0.58
unsp 552 WEIRD 0.45 0.42 0.48
others 108 Non-WEIRD 0.61 0.54 0.69
unsp 69 Non-WEIRD 0.39 0.32 0.47

[1] “Chi = 2.08 , p-value = 0.16791604197901”

Recruitment volunteers: [1] “volunteer n=368: 30% [27% to 32%]” [1] “volunteer n=56: 32% [25% to 39%]”

Volunteer samples
var n label prop lwr.ci upr.ci
others 866 WEIRD 0.70 0.68 0.73
volunteer 368 WEIRD 0.30 0.27 0.32
others 121 Non-WEIRD 0.68 0.62 0.75
volunteer 56 Non-WEIRD 0.32 0.25 0.39

[1] “Chi = 0.16 , p-value = 0.685221539615238”

Recruitment unspecified: [1] “Not specified n=483: 39% [36% to 42%]” [1] “Not specified n=90: 51% [44% to 59%]”

recruitment unspecified samples
var n label prop lwr.ci upr.ci
Not specified 483 WEIRD 0.39 0.36 0.42
others 751 WEIRD 0.61 0.58 0.64
Not specified 90 Non-WEIRD 0.51 0.44 0.59
others 87 Non-WEIRD 0.49 0.42 0.57

[1] “Chi = 8.32 , p-value = 0.00392766563342981”

Experimenter Created Music: [1] “experimenter n=278: 34% [31% to 38%]” [1] “experimenter n=45: 34% [27% to 43%]”

Experimenter selected. music
var n label prop lwr.ci upr.ci
other 532 WEIRD 0.66 0.62 0.69
experimenter 278 WEIRD 0.34 0.31 0.38
other 86 Non-WEIRD 0.66 0.58 0.74
experimenter 45 Non-WEIRD 0.34 0.27 0.43

[1] “Chi = 0 , p-value = 0.999999999999998”

Western music: [1] “western n=589: 73% [70% to 76%]” [1] “western n=74: 56% [48% to 65%]”

Western music
var n label prop lwr.ci upr.ci
other 221 WEIRD 0.27 0.24 0.30
western 589 WEIRD 0.73 0.70 0.76
other 57 Non-WEIRD 0.44 0.35 0.52
western 74 Non-WEIRD 0.56 0.48 0.65

[1] “Chi = 13.5 , p-value = 0.000238986429147995”

Music origin unspecified: [1] “Not specified n=189: 23% [20% to 26%]” [1] “Not specified n=28: 21% [15% to 29%]”

Music origin unspecified
var n label prop lwr.ci upr.ci
other 621 WEIRD 0.77 0.74 0.80
Not specified 189 WEIRD 0.23 0.20 0.26
other 103 Non-WEIRD 0.79 0.73 0.86
Not specified 28 Non-WEIRD 0.21 0.15 0.29

[1] “Chi = 0.15 , p-value = 0.70234703527474”

Descriptives for alternative WEIRD formulations: Cultural distance (from the US)

Muthukrishna et al., 2020, https://journals.sagepub.com/doi/full/10.1177/0956797620916782#supplementary-materials defines cultural distances between countries based on numerous indices (religion, political orientation, social, financial, sexual, law, media, etc.). Since this metric has been established for 80 countries, we inferred that some countries in our dataset are culturally or geographically proximate countries, listed below. See also http://www.culturaldistance.com

Austria from Germany
Beligum from France
Croatia from Slovenia
Czech Republic from Poland
Denmark from Sweden
Ireland from Great Britain
Portugal from Spain
Slovakia from Slovenia
Belgium from France
Central African Republic from Mali
Greece from Armenia
Iceland from Norway
Israel from Japan
Kenya from Ethiopia
Latvia from Estonia
Serbia from Slovenia
UAE from Malaysia

We split the countries into WEIRD or non-WEIRD based on mean distance from the US + 0.02.

source('scripts/alternative_WEIRD_indices.R')
d2 <- alternative_WEIRD_indices(d,'Muthukhrisna'); d <- d2$d; DF <- d2$DF

Using WEIRD country index: Muthukhrisna

knitr::kable(table(d$CountryDataCollected_WEOG)/sum(table(d$CountryDataCollected_WEOG)),digits = 2)
Var1 Freq
Non-WEOG 0.12
WEOG 0.88 
source('scripts/table1.R')
Sample Size
sample_country_data_collected_WEOG median lwr.ci upr.ci
WEOG 48.0 44.00 53.0
Non-WEOG 54.5 42.01 71.5

[1] “\(W = 103,461.00\), \(p = .486\)”

Age Mean
wtd.avg CI.LL CI.UL
Non-WEIRD 19.93 17.06 22.04
WEIRD 27.63 25.38 29.94

[1] “t value = 9.37, df = 203.71, p-value = 0”

Age SD:

Age SD
wtd.avg CI.LL CI.UL
Non-WEIRD 3.72 2.54 4.58
WEIRD 7.12 6.27 8.08

[1] “t value = 8.04, df = 151.22, p-value = 0”

Gender balance (country data collected):

Gender balance (Primary Country Data Collected)
wtd.avg CI.LL CI.UL
Non-WEIRD 0.54 0.49 0.56
WEIRD 0.59 0.57 0.61

[1] “t value = 4.17, df = 139.18, p-value = 0”

Gender balance (based on first author country):

Gender balance (Primary Author Country)
wtd.avg CI.LL CI.UL
Non-WEIRD 0.54 0.49 0.56
WEIRD 0.59 0.57 0.61

[1] “t value = 3.68, df = 109, p-value = 0”

Solely musicians:

Musical expertise across samples
Expertise %
musicians 30.5
musicians; non-musicians 22.0
non-musicians 10.1
Not specified 37.4

[1] “musicians n=381: 30% [28% to 33%]” [1] “musicians n=48: 30% [23% to 37%]”

Solely musicians
var n label prop lwr.ci upr.ci
musicians 381 WEIRD 0.3 0.28 0.33
others 870 WEIRD 0.7 0.67 0.72
musicians 48 Non-WEIRD 0.3 0.23 0.37
others 112 Non-WEIRD 0.7 0.63 0.77

[1] “Chi = 0.01 , p-value = 0.919540229885057”

musicians and non-musicians: [1] “musicians n=662: 53% [50% to 56%]” [1] “musicians n=80: 50% [42% to 58%]”

musicians and non-musicians
var n label prop lwr.ci upr.ci
musicians 662 WEIRD 0.53 0.50 0.56
non-musicians 589 WEIRD 0.47 0.44 0.50
musicians 80 Non-WEIRD 0.50 0.42 0.58
non-musicians 80 Non-WEIRD 0.50 0.42 0.58

[1] “Chi = 0.48 , p-value = 0.493753123438281”

Solely Non-musicians: [1] “non-musicians n=131: 10% [9% to 12%]” [1] “non-musicians n=22: 14% [9% to 19%]”

non-musicians
var n label prop lwr.ci upr.ci
others 1120 WEIRD 0.90 0.88 0.91
non-musicians 131 WEIRD 0.10 0.09 0.12
others 138 Non-WEIRD 0.86 0.82 0.92
non-musicians 22 Non-WEIRD 0.14 0.09 0.19

[1] “Chi = 1.58 , p-value = 0.222888555722139”

University sample: [1] “university n=510: 41% [38% to 44%]” [1] “university n=69: 43% [36% to 51%]”

University samples
var n label prop lwr.ci upr.ci
others 741 WEIRD 0.59 0.56 0.62
university 510 WEIRD 0.41 0.38 0.44
others 91 Non-WEIRD 0.57 0.49 0.65
university 69 Non-WEIRD 0.43 0.36 0.51

[1] “Chi = 0.33 , p-value = 0.59720139930035”

Sample unspecified: [1] “unsp n=563: 45% [42% to 48%]” [1] “unsp n=58: 36% [29% to 44%]”

Unspecified samples
var n label prop lwr.ci upr.ci
others 688 WEIRD 0.55 0.52 0.58
unsp 563 WEIRD 0.45 0.42 0.48
others 102 Non-WEIRD 0.64 0.57 0.72
unsp 58 Non-WEIRD 0.36 0.29 0.44

[1] “Chi = 4.41 , p-value = 0.039480259870065”

Recruitment volunteers: [1] “volunteer n=370: 30% [27% to 32%]” [1] “volunteer n=54: 34% [27% to 42%]”

Volunteer samples
var n label prop lwr.ci upr.ci
others 881 WEIRD 0.70 0.68 0.73
volunteer 370 WEIRD 0.30 0.27 0.32
others 106 Non-WEIRD 0.66 0.59 0.74
volunteer 54 Non-WEIRD 0.34 0.27 0.42

[1] “Chi = 0.99 , p-value = 0.320865528891456”

Recruitment unspecified: [1] “Not specified n=496: 40% [37% to 42%]” [1] “Not specified n=77: 48% [41% to 56%]”

recruitment unspecified samples
var n label prop lwr.ci upr.ci
Not specified 496 WEIRD 0.40 0.37 0.42
others 755 WEIRD 0.60 0.58 0.63
Not specified 77 Non-WEIRD 0.48 0.41 0.56
others 83 Non-WEIRD 0.52 0.44 0.60

[1] “Chi = 3.88 , p-value = 0.0488020308237839”

Experimenter Created Music: [1] “experimenter n=266: 34% [30% to 37%]” [1] “experimenter n=40: 38% [30% to 48%]”

Experimenter selected. music
var n label prop lwr.ci upr.ci
other 528 WEIRD 0.66 0.63 0.70
experimenter 266 WEIRD 0.34 0.30 0.37
other 65 Non-WEIRD 0.62 0.53 0.72
experimenter 40 Non-WEIRD 0.38 0.30 0.48

[1] “Chi = 0.68 , p-value = 0.409895772994663”

Western music: [1] “western n=585: 74% [71% to 77%]” [1] “western n=54: 51% [42% to 61%]”

Western music
var n label prop lwr.ci upr.ci
other 209 WEIRD 0.26 0.23 0.29
western 585 WEIRD 0.74 0.71 0.77
other 51 Non-WEIRD 0.49 0.39 0.58
western 54 Non-WEIRD 0.51 0.42 0.61

[1] “Chi = 21.26 , p-value = 4.00546262205008e-06”

Music origin unspecified: [1] “Not specified n=180: 23% [20% to 26%]” [1] “Not specified n=28: 27% [19% to 36%]”

Music origin unspecified
var n label prop lwr.ci upr.ci
other 614 WEIRD 0.77 0.74 0.80
Not specified 180 WEIRD 0.23 0.20 0.26
other 77 Non-WEIRD 0.73 0.66 0.82
Not specified 28 Non-WEIRD 0.27 0.19 0.36

[1] “Chi = 0.62 , p-value = 0.429797063240015”

Summary of differences

When these two alternative attributions (WEOG and Mutkukhrisna et al., 2020) of countries across WEIRD/non-WEIRD divide are incorporated into the analysis of the main trends in the sample of music psychology studies, the outcome reported in Table 1 with the country division provided by Krys et al. provides largely identical patterns despite small fluctuations in the WEIRD/non-WEIRD division (non-WEIRD in Krys et al. is 23%, in WEOG 16%, and in Mutkukhrisna 12% in this data).

To put the differences in the actual variables of interest more precisely, there are 11 statistical comparisons (of counts or means) in Table 1. For the division based on WEOG countries, one of the rows in Table 1 results in a difference (the sample size is significant at p<.05 level). For the second variant country division by Mutkukhrisna et al., there are no differences to the reported division of countries.

References

  • Krys, K., de Almeida, I., Wasiel, A., & Vignoles, V. L. (2024). WEIRD–Confucian comparisons: Ongoing cultural biases in psychology’s evidence base and some recommendations for improving global representation. American Psychologist. https://doi.org/10.1037/amp0001298

  • Muthukrishna, M., Bell, A. V., Henrich, J., Curtin, C. M., Gedranovich, A., McInerney, J., & Thue, B. (2020). Beyond Western, Educated, Industrial, Rich, and Democratic (WEIRD) Psychology: Measuring and Mapping Scales of Cultural and Psychological Distance. Psychological Science, 31(6), 678-701. https://doi.org/10.1177/0956797620916782

Back to top