loading libraries

library(psych)
library(report)

Reading in data

FinalData <- read.csv("wpa_lena_pos.csv")

Subsetting variables of interest

df <- FinalData[,1:8]

Aggregating sitting time variable

sit_mean <- aggregate(df$sit_time, by=list(df$id_uni), FUN=mean)
names(sit_mean)[names(sit_mean) == "Group.1"] <- "UniqueID"
names(sit_mean)[names(sit_mean) == "x"] <- "sit_ave"

Eliminating duplicates and merging data frames

df1 <- df[!duplicated(df$id_uni),]
df1.5 <- merge(sit_mean, df1, by.x = "UniqueID", by.y = "id_uni")

Aggregating held time variable

held_mean <- aggregate(df$held_time, by=list(df$id_uni), FUN=mean)
names(held_mean)[names(held_mean) == "Group.1"] <- "UniqueID"
names(held_mean)[names(held_mean) == "x"] <- "held_ave"

Merging data frames (df1.5 with held time data frame)

df2 <- merge(held_mean, df1.5, by.x = "UniqueID", by.y = "UniqueID")

Aggregating prone variable

prone_mean <- aggregate(df$prone_time, by=list(df$id_uni), FUN=mean)
names(prone_mean)[names(prone_mean) == "Group.1"] <- "UniqueID"
names(prone_mean)[names(prone_mean) == "x"] <- "prone_ave"

Merging data frames (df2 with held prone data frame)

df3 <- merge(prone_mean, df2, by.x = "UniqueID", by.y = "UniqueID")

Aggregating supine variable

supine_mean <- aggregate(df$supine_time, by=list(df$id_uni), FUN=mean)
names(supine_mean)[names(supine_mean) == "Group.1"] <- "UniqueID"
names(supine_mean)[names(supine_mean) == "x"] <- "supine_ave"

Merging data frames (df3 with held supine data frame)

df4 <- merge(supine_mean, df3, by.x = "UniqueID", by.y = "UniqueID")

Aggregating upright variable

upright_mean <- aggregate(df$upright_time, by=list(df$id_uni), FUN=mean)
names(upright_mean)[names(upright_mean) == "Group.1"] <- "UniqueID"
names(upright_mean)[names(upright_mean) == "x"] <- "upright_ave"

Merging data frames (df4 with upright data frame)

df5 <- merge(upright_mean, df4, by.x = "UniqueID", by.y = "UniqueID")

Aggregating adult word count variable

adult_word_count_mean <- aggregate(df$adult_word_cnt, by=list(df$id_uni), FUN=mean)
names(adult_word_count_mean)[names(adult_word_count_mean) == "Group.1"] <- "UniqueID"
names(adult_word_count_mean)[names(adult_word_count_mean) == "x"] <- "adult_word_count_ave"

Merging data frames (df5 with adult word count data frame)

df6 <- merge(adult_word_count_mean, df5, by.x = "UniqueID", by.y = "UniqueID")

Subsetting final variables of interest

df_final <- df6[,-c(9:14)]

Descriptive statistics on all variables

describe(df_final$sit_ave)
##    vars  n mean   sd median trimmed  mad min  max range  skew kurtosis   se
## X1    1 11 0.52 0.12   0.54    0.52 0.09 0.3 0.69  0.39 -0.28    -1.16 0.04
describe(df_final$held_ave)
##    vars  n mean   sd median trimmed  mad  min  max range skew kurtosis   se
## X1    1 11 0.08 0.05   0.06    0.07 0.05 0.02 0.16  0.14 0.66    -1.04 0.01
describe(df_final$prone_ave)
##    vars  n mean   sd median trimmed  mad  min  max range skew kurtosis   se
## X1    1 11 0.14 0.07   0.14    0.13 0.06 0.06 0.28  0.22 0.85    -0.44 0.02
describe(df_final$supine_ave)
##    vars  n mean   sd median trimmed mad  min max range  skew kurtosis   se
## X1    1 11 0.11 0.07   0.11    0.11 0.1 0.02 0.2  0.18 -0.09    -1.75 0.02
describe(df_final$upright_ave)
##    vars  n mean   sd median trimmed  mad  min  max range skew kurtosis   se
## X1    1 11 0.16 0.06   0.16    0.16 0.07 0.07 0.26  0.19 0.08    -1.25 0.02
describe(df_final$adult_word_count_ave)
##    vars  n mean   sd median trimmed   mad    min    max  range skew kurtosis
## X1    1 11  243 74.5  246.8  235.43 66.95 150.65 403.38 252.73  0.6    -0.52
##       se
## X1 22.46

correlation between time spent sitting and adult word count

correlation_sitting <- corr.test(df_final$sit_ave, df_final$adult_word_count_ave)
print(correlation_sitting, short = F)
## Call:corr.test(x = df_final$sit_ave, y = df_final$adult_word_count_ave)
## Correlation matrix 
## [1] 0.24
## Sample Size 
## [1] 11
## These are the unadjusted probability values.
##   The probability values  adjusted for multiple tests are in the p.adj object. 
## [1] 0.47
## 
##  Confidence intervals based upon normal theory.  To get bootstrapped values, try cor.ci
##       raw.lower raw.r raw.upper raw.p lower.adj upper.adj
## NA-NA     -0.42  0.24      0.73  0.47     -0.42      0.73
plot(df_final$sit_ave, df_final$adult_word_count_ave)

correlation between time spent held and adult word count

correlation_held <- corr.test(df_final$held_ave, df_final$adult_word_count_ave)
print(correlation_held, short = F)
## Call:corr.test(x = df_final$held_ave, y = df_final$adult_word_count_ave)
## Correlation matrix 
## [1] 0.11
## Sample Size 
## [1] 11
## These are the unadjusted probability values.
##   The probability values  adjusted for multiple tests are in the p.adj object. 
## [1] 0.75
## 
##  Confidence intervals based upon normal theory.  To get bootstrapped values, try cor.ci
##       raw.lower raw.r raw.upper raw.p lower.adj upper.adj
## NA-NA     -0.52  0.11      0.67  0.75     -0.52      0.67
plot(df_final$held_ave, df_final$adult_word_count_ave)

correlation between time spent prone and adult word count

correlation_prone <- corr.test(df_final$prone_ave, df_final$adult_word_count_ave)
print(correlation_prone, short = F)
## Call:corr.test(x = df_final$prone_ave, y = df_final$adult_word_count_ave)
## Correlation matrix 
## [1] 0.04
## Sample Size 
## [1] 11
## These are the unadjusted probability values.
##   The probability values  adjusted for multiple tests are in the p.adj object. 
## [1] 0.9
## 
##  Confidence intervals based upon normal theory.  To get bootstrapped values, try cor.ci
##       raw.lower raw.r raw.upper raw.p lower.adj upper.adj
## NA-NA     -0.57  0.04      0.63   0.9     -0.57      0.63
plot(df_final$prone_ave, df_final$adult_word_count_ave)

correlation between time spent supine and adult word count

correlation_supine <- corr.test(df_final$supine_ave, df_final$adult_word_count_ave)
print(correlation_supine, short = F)
## Call:corr.test(x = df_final$supine_ave, y = df_final$adult_word_count_ave)
## Correlation matrix 
## [1] -0.24
## Sample Size 
## [1] 11
## These are the unadjusted probability values.
##   The probability values  adjusted for multiple tests are in the p.adj object. 
## [1] 0.48
## 
##  Confidence intervals based upon normal theory.  To get bootstrapped values, try cor.ci
##       raw.lower raw.r raw.upper raw.p lower.adj upper.adj
## NA-NA     -0.73 -0.24      0.42  0.48     -0.73      0.42
plot(df_final$supine_ave, df_final$adult_word_count_ave)

correlation between time spent upright and adult word count

correlation_upright <- corr.test(df_final$upright_ave, df_final$adult_word_count_ave)
print(correlation_upright, short = F)
## Call:corr.test(x = df_final$upright_ave, y = df_final$adult_word_count_ave)
## Correlation matrix 
## [1] -0.34
## Sample Size 
## [1] 11
## These are the unadjusted probability values.
##   The probability values  adjusted for multiple tests are in the p.adj object. 
## [1] 0.31
## 
##  Confidence intervals based upon normal theory.  To get bootstrapped values, try cor.ci
##       raw.lower raw.r raw.upper raw.p lower.adj upper.adj
## NA-NA     -0.78 -0.34      0.33  0.31     -0.78      0.33
plot(df_final$upright_ave, df_final$adult_word_count_ave)

regression: time spent sitting as a predictor of adult word count

regression_sitting <- lm(adult_word_count_ave~sit_ave, data=df_final)
summary(regression_sitting)
## 
## Call:
## lm(formula = adult_word_count_ave ~ sit_ave, data = df_final)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -97.738 -39.603  -5.958  29.051 163.943 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)
## (Intercept)    164.3      107.8   1.525    0.162
## sit_ave        152.7      204.4   0.747    0.474
## 
## Residual standard error: 76.21 on 9 degrees of freedom
## Multiple R-squared:  0.05836,    Adjusted R-squared:  -0.04627 
## F-statistic: 0.5578 on 1 and 9 DF,  p-value: 0.4742
report(regression_sitting)
## We fitted a linear model (estimated using OLS) to predict adult_word_count_ave
## with sit_ave (formula: adult_word_count_ave ~ sit_ave). The model explains a
## statistically not significant and weak proportion of variance (R2 = 0.06, F(1,
## 9) = 0.56, p = 0.474, adj. R2 = -0.05). The model's intercept, corresponding to
## sit_ave = 0, is at 164.34 (95% CI [-79.51, 408.19], t(9) = 1.52, p = 0.162).
## Within this model:
## 
##   - The effect of sit ave is statistically non-significant and positive (beta =
## 152.67, 95% CI [-309.76, 615.11], t(9) = 0.75, p = 0.474; Std. beta = 0.24, 95%
## CI [-0.49, 0.97])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald t-distribution approximation.

regression: time spent held as a predictor of adult word count

regression_held <- lm(adult_word_count_ave~held_ave, data=df_final)
summary(regression_held)
## 
## Call:
## lm(formula = adult_word_count_ave ~ held_ave, data = df_final)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -105.872  -45.105    1.572   31.289  156.894 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    229.2       47.4   4.836 0.000926 ***
## held_ave       176.0      526.4   0.334 0.745743    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 78.05 on 9 degrees of freedom
## Multiple R-squared:  0.01227,    Adjusted R-squared:  -0.09748 
## F-statistic: 0.1118 on 1 and 9 DF,  p-value: 0.7457
report(regression_held)
## We fitted a linear model (estimated using OLS) to predict adult_word_count_ave
## with held_ave (formula: adult_word_count_ave ~ held_ave). The model explains a
## statistically not significant and very weak proportion of variance (R2 = 0.01,
## F(1, 9) = 0.11, p = 0.746, adj. R2 = -0.10). The model's intercept,
## corresponding to held_ave = 0, is at 229.24 (95% CI [122.01, 336.47], t(9) =
## 4.84, p < .001). Within this model:
## 
##   - The effect of held ave is statistically non-significant and positive (beta =
## 176.02, 95% CI [-1014.76, 1366.80], t(9) = 0.33, p = 0.746; Std. beta = 0.11,
## 95% CI [-0.64, 0.86])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald t-distribution approximation.

regression: time spent prone as a predictor of adult word count

regression_prone <- lm(adult_word_count_ave~prone_ave, data=df_final)
summary(regression_prone)
## 
## Call:
## lm(formula = adult_word_count_ave ~ prone_ave, data = df_final)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -92.846 -51.598   3.714  37.921 155.758 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)   
## (Intercept)   236.14      57.04   4.140  0.00252 **
## prone_ave      49.80     377.22   0.132  0.89788   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 78.46 on 9 degrees of freedom
## Multiple R-squared:  0.001933,   Adjusted R-squared:  -0.109 
## F-statistic: 0.01743 on 1 and 9 DF,  p-value: 0.8979
report(regression_prone)
## We fitted a linear model (estimated using OLS) to predict adult_word_count_ave
## with prone_ave (formula: adult_word_count_ave ~ prone_ave). The model explains
## a statistically not significant and very weak proportion of variance (R2 =
## 1.93e-03, F(1, 9) = 0.02, p = 0.898, adj. R2 = -0.11). The model's intercept,
## corresponding to prone_ave = 0, is at 236.14 (95% CI [107.10, 365.19], t(9) =
## 4.14, p = 0.003). Within this model:
## 
##   - The effect of prone ave is statistically non-significant and positive (beta =
## 49.80, 95% CI [-803.53, 903.13], t(9) = 0.13, p = 0.898; Std. beta = 0.04, 95%
## CI [-0.71, 0.80])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald t-distribution approximation.

regression: time spent supine as a predictor of adult word count

regression_supine <- lm(adult_word_count_ave~supine_ave, data=df_final)
summary(regression_supine)
## 
## Call:
## lm(formula = adult_word_count_ave ~ supine_ave, data = df_final)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -113.36  -35.28  -11.52   33.60  160.12 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    272.5       45.9   5.938 0.000219 ***
## supine_ave    -266.1      358.0  -0.743 0.476319    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 76.23 on 9 degrees of freedom
## Multiple R-squared:  0.05782,    Adjusted R-squared:  -0.04687 
## F-statistic: 0.5523 on 1 and 9 DF,  p-value: 0.4763
report(regression_supine)
## We fitted a linear model (estimated using OLS) to predict adult_word_count_ave
## with supine_ave (formula: adult_word_count_ave ~ supine_ave). The model
## explains a statistically not significant and weak proportion of variance (R2 =
## 0.06, F(1, 9) = 0.55, p = 0.476, adj. R2 = -0.05). The model's intercept,
## corresponding to supine_ave = 0, is at 272.52 (95% CI [168.69, 376.35], t(9) =
## 5.94, p < .001). Within this model:
## 
##   - The effect of supine ave is statistically non-significant and negative (beta
## = -266.07, 95% CI [-1075.93, 543.80], t(9) = -0.74, p = 0.476; Std. beta =
## -0.24, 95% CI [-0.97, 0.49])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald t-distribution approximation.

regression: time spent upright as a predictor of adult word count

regression_upright <- lm(adult_word_count_ave~upright_ave, data=df_final)
summary(regression_upright)
## 
## Call:
## lm(formula = adult_word_count_ave ~ upright_ave, data = df_final)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -123.429  -51.857    5.907   42.177  123.975 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)   
## (Intercept)   307.99      64.83   4.751  0.00104 **
## upright_ave  -411.15     385.06  -1.068  0.31344   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 73.99 on 9 degrees of freedom
## Multiple R-squared:  0.1124, Adjusted R-squared:  0.01381 
## F-statistic:  1.14 on 1 and 9 DF,  p-value: 0.3134
report(regression_upright)
## We fitted a linear model (estimated using OLS) to predict adult_word_count_ave
## with upright_ave (formula: adult_word_count_ave ~ upright_ave). The model
## explains a statistically not significant and weak proportion of variance (R2 =
## 0.11, F(1, 9) = 1.14, p = 0.313, adj. R2 = 0.01). The model's intercept,
## corresponding to upright_ave = 0, is at 307.99 (95% CI [161.34, 454.63], t(9) =
## 4.75, p = 0.001). Within this model:
## 
##   - The effect of upright ave is statistically non-significant and negative (beta
## = -411.15, 95% CI [-1282.23, 459.93], t(9) = -1.07, p = 0.313; Std. beta =
## -0.34, 95% CI [-1.05, 0.38])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald t-distribution approximation.