These are subsets from Vietnam SUNRISE pilot study data.
TPAaccelerometer: TPA time measured by Actigraph accelerometer (mins)
TPAquestionnaire: TPA tim measured by self-administered parent questionnaire (mins)
Average = (TPAaccelerometer + TPAquestionnaire)/2 (mins)
Bias = TPAaccelerometer - TPAquestionnaire (mins)
## 'data.frame': 112 obs. of 4 variables:
## $ TPAaccelerometer: num 202 159 212 120 177 ...
## $ TPAquestionnaire: int 240 480 120 NA 240 330 120 120 480 10 ...
## $ average : num 221 320 166 NA 209 ...
## $ bias : num -38.2 -320.6 92.5 NA -62.5 ...
## TPAaccelerometer TPAquestionnaire average bias
## Min. :101.8 Min. : 10.0 Min. : 89.54 Min. :-340.083
## 1st Qu.:152.7 1st Qu.: 97.5 1st Qu.:142.75 1st Qu.: -86.875
## Median :181.5 Median :140.0 Median :169.41 Median : 45.750
## Mean :185.8 Mean :197.0 Mean :192.03 Mean : -9.921
## 3rd Qu.:212.9 3rd Qu.:270.0 3rd Qu.:227.40 3rd Qu.: 103.750
## Max. :310.8 Max. :480.0 Max. :359.88 Max. : 245.625
## NA's :14 NA's :14 NA's :14
Mean of TPA time by Acceleremeter: 185.8 mins
Mean of TPA time by Parent Questionnaire: 196.99 mins
T test result (unequal variance)
##
## Welch Two Sample t-test
##
## data: compare2$TPAaccelerometer and compare2$TPAquestionnaire
## t = -0.75268, df = 112.36, p-value = 0.4532
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -40.64655 18.26621
## sample estimates:
## mean of x mean of y
## 185.7996 196.9898
Data from Vietnam SUNRISE pilot study shows that time parents did not underreport children’s PA levels before COVID19.
Indeed, time reported via questionnaire were higher than time measured via accelerometer, although the effect was small.
Also, Bland-Altman analysis shows that the parent questionnaire
Underestimate the PA time when a child’s TPA was around 180 - 200 mins/day.
Overestimate the PA time when a child’s TPA was higher than around 180 - 200 mins/day.