For this exercise, please try to reproduce the results from Experiment 1 of the associated paper (Ko, Sadler & Galinsky, 2015). The PDF of the paper is included in the same folder as this Rmd file.
A sense of power has often been tied to how we perceive each other’s voice. Social hierarchy is embedded into the structure of society and provides a metric by which others relate to one another. In 1956, the Brunswik Lens Model was introduced to examine how vocal cues might influence hierarchy. In “The Sound of Power: Conveying and Detecting Hierarchical Rank Through Voice,” Ko and colleagues investigated how manipulation of hierarchal rank within a situation might impact vocal acoustic cues. Using the Brunswik Model, six acoustic metrics were utilized (pitch mean & variability, loudness mean & variability, and resonance mean & variability) to isolate a potential contribution between individuals of different hierarchal rank. In the first experiment, Ko, Sadler & Galinsky examined the vocal acoustic cues of individuals before and after being assigned a hierarchal rank in a sample of 161 subjects (80 male). Each of the six hierarchy acoustic cues were analyzed with a 2 (high vs. low rank condition) x 2 (male vs. female) analysis of covariance, controlling for the baseline of the respective acoustic cue.
Below is the specific result you will attempt to reproduce (quoted directly from the results section of Experiment 1):
The impact of hierarchical rank on speakers’ acoustic cues. Each of the six hierarchy-based (i.e., postmanipulation) acoustic variables was submitted to a 2 (condition: high rank, low rank) × 2 (speaker’s sex: female, male) between-subjects analysis of covariance, controlling for the corresponding baseline acoustic variable. Table 4 presents the adjusted means by condition. Condition had a significant effect on pitch, pitch variability, and loudness variability. Speakers’ voices in the high-rank condition had higher pitch, F(1, 156) = 4.48, p < .05; were more variable in loudness, F(1, 156) = 4.66, p < .05; and were more monotone (i.e., less variable in pitch), F(1, 156) = 4.73, p < .05, compared with speakers’ voices in the low-rank condition (all other Fs < 1; see the Supplemental Material for additional analyses of covariance involving pitch and loudness). (from Ko et al., 2015, p. 6; emphasis added)
The adjusted means for these analyses are reported in Table 4 (Table4_AdjustedMeans.png, included in the same folder as this Rmd file).
library(tidyverse) # for data munging
library(knitr) # for kable table formating
library(haven) # import and export 'SPSS', 'Stata' and 'SAS' Files
library(readxl) # import excel files
library(emmeans)
# #optional packages:
# library(psych)
library(car) # for ANCOVA
# library(compute.es) # for ANCOVA
# library(lsmeans) # for ANCOVA# Just Experiment 1
d <-read_csv("data/S1_voice_level_Final.csv")
# DT::datatable(d)# Having a hard time tidying the data since I'm having difficulty understanding what the variables in the df are# Same as above -- hard to preprocess when I don't understand the structure of the dfIn the paper, the adjusted means by condition are reported (see Table 4, or Table4_AdjustedMeans.png, included in the same folder as this Rmd file). Reproduce these values below:
# Pitch and pitch variability
pitch <- lm(pitch_rmean ~ plev + pitch_smean, data = d)
emmeans(pitch, ~ plev)## plev emmean SE df lower.CL upper.CL
## -1 151 0.987 158 149 153
## 1 148 0.969 158 146 150
##
## Confidence level used: 0.95pitch_variability <- lm(pitch_rvar ~ plev + pitch_svar, data = d)
emmeans(pitch_variability, ~ plev)## plev emmean SE df lower.CL upper.CL
## -1 1686 118 158 1454 1919
## 1 1816 115 158 1588 2044
##
## Confidence level used: 0.95# Loudness and loudness variability
loudness <- lm(intense_rmean ~ plev + intense_smean, data = d)
emmeans(loudness, ~ plev)## plev emmean SE df lower.CL upper.CL
## -1 57.6 0.331 158 57.0 58.3
## 1 57.3 0.325 158 56.6 57.9
##
## Confidence level used: 0.95loudness_variability <- lm(intense_rvar ~ plev + intense_svar, data = d)
emmeans(loudness_variability, ~ plev)## plev emmean SE df lower.CL upper.CL
## -1 186 4.71 158 177 195
## 1 180 4.63 158 171 189
##
## Confidence level used: 0.95# Resonance and resonance variability
resonance <- lm(form_rmean ~ plev + form_smean, data = d)
emmeans(resonance, ~ plev)## plev emmean SE df lower.CL upper.CL
## -1 1295 4.81 158 1286 1305
## 1 1290 4.72 158 1281 1300
##
## Confidence level used: 0.95resonance_variability <- lm(form_rvar ~ plev + form_svar, data = d)
emmeans(resonance_variability, ~ plev)## plev emmean SE df lower.CL upper.CL
## -1 64986 1440 158 62144 67828
## 1 63306 1410 158 60517 66096
##
## Confidence level used: 0.95The impact of hierarchical rank on speakers’ acoustic cues. Each of the six hierarchy-based (i.e., postmanipulation) acoustic variables was submitted to a 2 (condition: high rank, low rank) × 2 (speaker’s sex: female, male) between-subjects analysis of covariance, controlling for the corresponding baseline acoustic variable. […] Condition had a significant effect on pitch, pitch variability, and loudness variability. Speakers’ voices in the high-rank condition had higher pitch, F(1, 156) = 4.48, p < .05; were more variable in loudness, F(1, 156) = 4.66, p < .05; and were more monotone (i.e., less variable in pitch), F(1, 156) = 4.73, p < .05, compared with speakers’ voices in the low-rank condition (all other Fs < 1; see the Supplemental Material for additional analyses of covariance involving pitch and loudness).
# reproduce the above results here
# Pitch and pitch variability
pitch_ancova <- aov(pitch_rmean ~ plev*vsex + pitch_smean, data = d)
anova(pitch_ancova)## Analysis of Variance Table
##
## Response: pitch_rmean
## Df Sum Sq Mean Sq F value Pr(>F)
## plev 1 1151 1151 16.3991 8.062e-05 ***
## vsex 1 217478 217478 3099.7853 < 2.2e-16 ***
## pitch_smean 1 36036 36036 513.6315 < 2.2e-16 ***
## plev:vsex 1 285 285 4.0688 0.0454 *
## Residuals 156 10945 70
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1pitch_variance_ancova <- aov(pitch_rvar ~ plev*vsex + pitch_svar, data = d)
anova(pitch_variance_ancova)## Analysis of Variance Table
##
## Response: pitch_rvar
## Df Sum Sq Mean Sq F value Pr(>F)
## plev 1 229785 229785 0.2164 0.642412
## vsex 1 7768624 7768624 7.3175 0.007588 **
## pitch_svar 1 70659662 70659662 66.5569 1.057e-13 ***
## plev:vsex 1 559469 559469 0.5270 0.468965
## Residuals 156 165616319 1061643
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Loudness and loudness variability
intensity_ancova <- aov(intense_rmean ~ plev*vsex + intense_smean, data = d)
anova(intensity_ancova)## Analysis of Variance Table
##
## Response: intense_rmean
## Df Sum Sq Mean Sq F value Pr(>F)
## plev 1 0.57 0.57 0.0757 0.7835
## vsex 1 174.41 174.41 23.1890 3.446e-06 ***
## intense_smean 1 1126.87 1126.87 149.8272 < 2.2e-16 ***
## plev:vsex 1 0.03 0.03 0.0034 0.9537
## Residuals 156 1173.30 7.52
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1intensity_variance_ancova <- aov(intense_rvar ~ plev*vsex + intense_svar, data = d)
anova(intensity_variance_ancova)## Analysis of Variance Table
##
## Response: intense_rvar
## Df Sum Sq Mean Sq F value Pr(>F)
## plev 1 787 787 0.5005 0.48035
## vsex 1 8538 8538 5.4299 0.02108 *
## intense_svar 1 192760 192760 122.5868 < 2e-16 ***
## plev:vsex 1 300 300 0.1906 0.66304
## Residuals 156 245300 1572
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Resonance and resonance variability
resonance_ancova <- aov(form_rmean ~ plev*vsex + form_smean, data = d)
anova(resonance_ancova)## Analysis of Variance Table
##
## Response: form_rmean
## Df Sum Sq Mean Sq F value Pr(>F)
## plev 1 1196 1196.2 0.6499 0.4213675
## vsex 1 6162 6162.3 3.3480 0.0691942 .
## form_smean 1 23517 23517.0 12.7771 0.0004673 ***
## plev:vsex 1 405 404.7 0.2199 0.6398048
## Residuals 156 287128 1840.6
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1resonance_variance_ancova <- aov(form_rvar ~ plev*vsex + form_svar, data = d)
anova(resonance_variance_ancova)## Analysis of Variance Table
##
## Response: form_rvar
## Df Sum Sq Mean Sq F value Pr(>F)
## plev 1 1.4407e+08 144072976 1.0917 0.2977
## vsex 1 5.1305e+09 5130509413 38.8774 4.042e-09 ***
## form_svar 1 3.4748e+08 347482476 2.6331 0.1067
## plev:vsex 1 1.2525e+08 125251423 0.9491 0.3315
## Residuals 156 2.0587e+10 131966320
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Were you able to reproduce the results you attempted to reproduce? If not, what part(s) were you unable to reproduce?
I had a much harder time with reproducing these results than the one I chose for Group A. I got somewhat close to reproducing the descriptive statistics and was able to find some of the same things significant for the inferential statistics, but that was about it.
How difficult was it to reproduce your results?
I had a hard time with this one – I ran into the 3 hour time limit.
What aspects made it difficult? What aspects made it easy?
Without a codebook, I wasn’t clear on what all the variables were. I didn’t also understand all the details of their analysis. Like when they said that they “[controlled] for the baseline of the respective acoustic cue” I wasn’t sure exactly what that meant. I didn’t really know how to implement that in my analysis. Perhaps that’s a common thing to do in this kind of research, and if I worked in this area it would be obvious. But for me, it wasn’t.