Package 'nipnTK'

Title: National Information Platforms for Nutrition Anthropometric Data Toolkit
Description: An implementation of the National Information Platforms for Nutrition or NiPN's analytic methods for assessing quality of anthropometric datasets that include measurements of weight, height or length, middle upper arm circumference, sex and age. The focus is on anthropometric status but many of the presented methods could be applied to other variables.
Authors: Mark Myatt [aut] , Ernest Guevarra [aut, cre]
Maintainer: Ernest Guevarra <[email protected]>
License: GPL-3
Version: 0.2.0.9000
Built: 2025-01-27 05:26:30 UTC
Source: https://github.com/nutriverse/nipnTK

Help Index

Goodness of fit to an expected model-based age distribution

Description

A simple model-based method for calculating expected numbers using exponential decay in a population in which births and deaths balance each other and with a 1:1 male to female sex ratio. This function is built specifically to test goodness of fit for a sample of children aged 6-59 months old grouped into four 1 year age groups and 1 half year age group (6 to less than 18 months, 18 to less than 30 months, 30 to less than 42 months, 42 to less than 54 months, and 54 months to less than 60 months).

Usage

ageChildren(age, u5mr = 1)

Arguments

age

A vector of ages. Should either be in whole months (integer) or in calculated decimal months (numeric).
u5mr

A numeric value for under five years mortality rate expressed as deaths / 10,000 persons / day. Default is set to 1.

Value

A list of class "ageChildren" with:

Variable Description
u5mr Under five years mortality rate as deaths / 10000 persons / day
observed Table of counts in each (year-centred) age group
expected Table of expected counts in each (year-centred) age group
X2 Chi-squared test statistic
df Degrees of freedom for Chi-squared test
p p-value for Chi-squared test

Examples

# Chi-Squared test for age of children in dp.ex02 sample dataset using an
# u5mr of 1 / 10,000 / day.
svy <- dp.ex02
ac <- ageChildren(svy$age, u5mr = 1)
ac

# Apply function to each sex separately
# Males
acM <- ageChildren(svy$age[svy$sex == 1], u5mr = 1)
acM
# Females
acF <- ageChildren(svy$age[svy$sex == 2], u5mr = 1)

# Simplified call to function by sex
by(svy$age, svy$sex, ageChildren, u5mr = 1)

Age-heaping analysis

Description

Age heaping is the tendency to report children's ages to the nearest year or adults’ ages to the nearest multiple of five or ten years. Age heaping is very common. This is a major reason why data from nutritional anthropometry surveys is often analysed and reported using broad age groups.

Usage

ageHeaping(x, divisor = 12)

Arguments

x

A vector of ages. Should either be in whole months (integer) or in calculated decimal months (numeric).
divisor

Divisor (usually 5, 6, 10, or 12); default is 12

Value

A list of class "ageHeaping" with:

Variable Description
X2 Chi-squared test statistic
df Degrees of freedom or Chi-squared test
p p-value for Chi-squared test
tab Table of remainders (for ⁠x \%\%⁠ divisor)
pct Table of proportions (⁠\%⁠) of remainders for ⁠x \%\%⁠ divisor)

Examples

# Test for age heaping using SMART survey data in Kabul, Afghanistan (dp.ex02)
# using a divisor of 12
svy <- dp.ex02
ah12 <- ageHeaping(svy$age)
ah12

# Test for age heaping using SMART survey data in Kabul, Afthanistan (dp.ex02)
# using a divisor of 6
ah6 <- ageHeaping(svy$age, divisor = 6)
ah6

Age ratio test

Description

Age ratio test is an age-related test of survey and data quality. In this test, the ratio of the number of children aged from 6 to 29 months to the number of children aged from 30 to 59 months is calculated. This ratio is then compared to an expected ratio (usually set at 0.85). The difference of the observed ratio to the expected ratio is then compared statistically using Chi-squared test.

Usage

ageRatioTest(x, ratio = 0.85)

Arguments

x

A vector of ages. Should either be in whole months (integer) or in calculated decimal months (numeric).
ratio

Expected age ratio. Default is 0.85.

Value

A lit of class "ageRatioTest" with:

Variable Description
expectedR Expected sex ratio
expectedP Expected proportion aged 6:29 months
observedR Observed sex ratio
observedP Observed proportion aged 6:29 months
X2 Chi-squared test statistic
df Degrees of freedom for Chi-squared test
p p-value for Chi-squared test

Examples

# Age-ratio test on survey dataset from Kabul, Afghanistan (`dp.ex02`)
# with an age ratio of 0.85
svy <- dp.ex02
ageRatioTest(svy$age, ratio = 0.85)

# The age ratio test applied to data for each sex separately
by(svy$age, svy$sex, ageRatioTest, ratio = 0.85)

Example dataset for age heaping function

Description

Anthropometric data from a Rapid Assessment Method for Older People (RAM-OP) survey in the Dadaab refugee camp in Garissa, Kenya. This is a survey of people aged sixty years and older.

Usage

ah.ex01

Format

A data frame with 593 observations and 10 variables

Variable Description
psu Primary sampling unit
camp Camp name code
block Block code
age Age (years)
sex Sex
weight Weight (kg)
height Height (cm)
demispan Demispan (cm)
muac Mid-upper arm circumference (cm)
oedema Oedema

Source

Data courtesy of HelpAge International

Example dataset for age and sex distributions function

Description

Data taken from household rosters collected as part of a household survey in Tanzania.

Usage

as.ex01

Format

A data frame of 8736 observations and 2 variables

Variable Description
age Age (years)
sex Sex (1 = Male / 2 = Female)

Example dataset for age and sex distributions function

Description

Census data of Tanzania taken from the Wolfram|Alpha knowledge engine.

Usage

as.ex02

Format

A data frame with 20 observations and 4 variables

Variable Description
age Age group
Males Total male population
Females Total female population
All Total population

Source

http://www.wolframalpha.com/input/?i=Tanzania+age+distribution

Plot text in a coloured bounding box.

Description

Plot text in a coloured bounding box.

Usage

boxText(
  x,
  y,
  labels,
  cex = 0.75,
  col = "white",
  border = FALSE,
  lwd = 0.5,
  pad = TRUE
)

Arguments

x, y

Co-ordinates of text that is to be plotted
labels

Text to be plotted
cex

Character expansion
col

Background colour
border

Border colour
lwd

Border width
pad

Add padding to (L) and (R) ends of bounding box

Examples

## Use of boxtext in the ageHeaping plot function
svy <- dp.ex02
ah12 <- ageHeaping(svy$age)

plot.new()
boxText(x = as.numeric(names(ah12$tab)),
        y = max(ah12$tab) * 0.1,
        labels = paste(sprintf(fmt = "%3.1f", ah12$pct), "%", sep = ""),
        cex = 0.5,
        pad = TRUE)

Digit preference test

Description

Digit preference is the observation that the final number in a measurement occurs with a greater frequency that is expected by chance. This can occur because of rounding, the practice of increasing or decreasing the value in a measurement to the nearest whole or half unit, or because data are made up. The digitPreference() function assesses the level by which digit preference exists in a given dataset using a ⁠digit preference score (DPS)⁠.

Usage

digitPreference(x, digits = 1, values = 0:9)

Arguments

x

Numeric vector of measurements
digits

Number of decimal places in x. Using digits = 1 (e.g.) allows 105 to be treated as 105.0
values

A vector of possible values for the final digit (default = 0:9)

Details

DPS definition from:

Kari Kuulasmaa K, Hense HW, Tolonen H (for the WHO MONICA Project), Quality Assessment of Data on Blood Pressure in the WHO MONICA Project, WHO MONICA Project e-publications No. 9, WHO, Geneva, May 1998 available from https://www.thl.fi/publications/monica/bp/bpqa.htm

Value

A list of class "digitPreference" with:

Variable Description
dps Digit Preference Score (DPS)
tab Table of final digit counts
pct Table of proportions (\%) of final digit counts

Examples

# Digit preference test applied to anthropometric data from a single state
# from a DHS survey in a West African country
svy <- dp.ex01
digitPreference(svy$wt, digits = 1)

Example dataset for distributions of variables and indices

Description

Anthropometric data from a SMART survey in Kabul, Afghanistan.

Usage

dist.ex01

Format

A data frame with 873 observations and 11 variables

Variable Description
psu Primary sampling unit
age Age of child (months)
sex Gender of child
weight Weight of child (kgs)
height Height of child (cm)
muac Mid-upper arm circumference (mm)
oedema Presence or absence of oedema
haz Height-for-age z-score
waz Weight-for-age z-score
whz Weight-for-height z-score
flag Data quality flag

Example dataset for digit preference function

Description

Anthropometric data from a single state from a Demographic and Health Survey (DHS) of a West African country.

Usage

dp.ex01

Format

A data frame with 796 observations and 6 variables

Variable Description
psu Primary sampling unit
age Age (months)
sex Gender
wt Weight (kg)
ht height (cm)
oedema Presence or absence of oedema

Example dataset for digit preference function

Description

Anthropometric data from a SMART survey in Kabul, Afghanistan in a comma-separated-value (CSV) file format. This is a survey of children aged 6-59 months old.

Usage

dp.ex02

Format

A data frame with 873 observations and 7 variables

Variable Description
psu Primary sampling unit
age Age of child (months)
sex Gender of child
weight Weight of child (kgs)
height Height of child (cm)
muac Mid-upper arm circumference (mm)
oedema Presence or absence of oedema

Example dataset for digit preference

Description

Anthropometric data for a sample of children living in a refugee camp in a West African country.

Usage

dp.ex03

Format

A data frame with 374 observations and 6 variables

Variable Description
age Age (months)
sex Gender (1 = Male / 2 = Female)
weight Weight (kg)
height Height (cm)
muac Mid-upper arm circumference (mm)
oedema Presence or absence of oedema

Example dataset for identifying outliers using flags

Description

Anthropometric data from a SMART survey in Sudan.

Usage

flag.ex01

Format

A data frame with 786 observations and 11 variables

Variable Description
psu Primary sampling unit
child Child ID
age Age (months)
sex Gender (1 = Male / 2 = Female)
weight Weight (kg)
height Height (cm)
muac Mid-upper arm circumference (mm)
oedema Presence or absence of oedema
haz Height-for-age z-score
waz Weight-for-age z-score
whz Weight-for-height z-score

Example dataset for identifying outliers using flags

Description

Anthropometric data from a survey of children 11 years or older attending school in Ethiopia.

Usage

flag.ex02

Format

A data.frame with 973 observations and 7 variables.

Variable Description
school School ID
sex Gender (1 = Male / 2 = Female)
ageMonths Age (months)
weight Weight (kg)
height Height (cm)
haz Height-for-age z-score
baz Body mass index (BMI)-for-age z-score

Example dataset for identifying outliers using flags

Description

Anthropometric data from a national survey in Nigeria.

Usage

flag.ex03

Format

A data frame with 18330 observations and 10 variables

Variable Description
psu Primary sampling unit
region Region code
state State
age Age (months)
sex Gender (1 = Male / 2 = Female)
weight Weight (kg)
height Height (cm)
haz Height-for-age z-score
waz Weight-for-age z-score
whz Weight-for-height z-score

Fill out a one-dimensional table to include a specified range of values

Description

Fill out a one-dimensional table to include a specified range of values

Usage

fullTable(x, values = min(x, na.rm = TRUE):max(x, na.rm = TRUE))

Arguments

x

A vector to tabulate
values

A vector of values to be included in a table. Default is: min(x, na.rm = TRUE):max(x, na.rm = TRUE)

Value

A table object including zero cells

Examples

# Generate some artificial data and then apply `fullTable()`
set.seed(0)
finalDigits <- sample(x = 0:9, size = 1000, replace = TRUE)
fullTable(finalDigits)

Green's Index of Dispersion

Description

Implementation of the Green's Index of Dispersion by bootstrap. The sampling distribution of the Green's Index is not well described hence bootstrapping is used to test whether the distribution of cases across primary sampling units is random.

Usage

greensIndex(data, psu, case, replicates = 999)

Arguments

data

Survey dataset (as an R data.frame)
psu

Name of variable holding PSU (cluster) data as a character vector of length = 1 (e.g. psu)
case

Name of variable holding case status as a character vector of length = 1 (e.g. GAM). The function assumes that case status is coded with 1 = case
replicates

Number of bootstrap replicates (default is 9999)

Details

The value of Green's Index can range between -1/(n - 1) for maximum uniformity (specific to the dataset) and one for maximum clumping. The interpretation of Green’s Index is straightforward:

Green's Index Value Interpretation
Green's Index close to 0 Random
Green's Index greater than 0 Clumped (i.e. more clumped than random)
Green’s Index less than 0 Uniform (i.e. more uniform than random)

Value

A list of class GI with names:

Variable Description
GI Estimate of Green's index
LCL 95\% LCL for GI
UCL 95\% UCL for GI
minGI Minimum possible GI (maximum uniformity) for the data
p p-value (H0: = Random distribution of cases across PSUs)

Examples

# Apply Green's Index using anthropometric data from a SMART survey in Sudan
# (flag.ex01)
svy <- flag.ex01
svy$flag <- 0
svy$flag <- ifelse(!is.na(svy$haz) & (svy$haz < -6 | svy$haz > 6),
                   svy$flag + 1, svy$flag)
svy$flag <- ifelse(!is.na(svy$whz) & (svy$whz < -5 | svy$whz > 5),
                   svy$flag + 2, svy$flag)
svy$flag <- ifelse(!is.na(svy$waz) & (svy$waz < -6 | svy$waz > 5),
                   svy$flag + 4, svy$flag)
svy <- svy[svy$flag == 0, ]
svy$stunted <- ifelse(svy$haz < -2, 1, 2)

## set seed to 0 to replicate results
set.seed(0)
greensIndex(data = svy, psu = "psu", case = "stunted")

Histogram with normal curve superimposed to help with “by-eye” assessments of normality of distribution

Description

Histogram with normal curve superimposed to help with “by-eye” assessments of normality of distribution

Usage

histNormal(
  x,
  xlab = deparse(substitute(x)),
  ylab = "Frequency",
  main = deparse(substitute(x)),
  breaks = "Sturges",
  ylim = NULL
)

Arguments

x

A numeric vector
xlab

x-axis label
ylab

y-axis label
main

Plot title
breaks

Passed to hist() function (?hist for details)
ylim

y-axis limits

Examples

# histNormal() with data from a SMART survey in Kabul, Afghanistan
# (dist.ex01)
svy <- dist.ex01
histNormal(svy$muac)
histNormal(svy$haz)
histNormal(svy$waz)
histNormal(svy$whz)

Add SMART flags to a stratified sample survey (e.g. MICS, DHS, national SMART)

Description

Add SMART flags to a stratified sample survey (e.g. MICS, DHS, national SMART)

Usage

national.SMART(x, strata, indices = c("haz", "whz", "waz"))

Arguments

x

Survey dataset (as an R data.frame) with indices present
strata

Name of column in x that defines the strata
indices

Names of columns in x containing indices

Value

A data.frame with same structure as x with a flagSMART column added. This column is coded using sums of powers of two

Examples

# Use the national.SMART() function to flag indices from a national
# SMART survey in Nigeria (flag.ex03)
svy <- flag.ex03
svyFlagged <- national.SMART(x = svy, strata = "state")

# Exclude records with flagging codes relevant to whz:
svyFlagged <- svyFlagged[!(svyFlagged$flagSMART %in% c(2, 3, 6, 7)), ]

Mahalanobis distance to detect bivariate outliers

Description

Mahalanobis distance to detect bivariate outliers

Usage

outliersMD(x, y, alpha = 0.001)

Arguments

x

Numeric vector
y

Numeric vector
alpha

Critical alpha value to detect an outlier (defaults to 0.001)

Value

A logical vector (TRUE for an outlier at p < alpha)

Examples

# Use outliersMD() to detect outliers in an anthropometric data from
# a SMART survey from the Democratic Republic of Congo (sp.ex01)
svy <- sp.ex01
svy[outliersMD(svy$height,svy$weight), ]

IQR to detect univariate outliers

Description

IQR to detect univariate outliers

Usage

outliersUV(x, fence = 1.5)

Arguments

x

Numeric vector
fence

IQR multiplier (defaults to 1.5)

Value

A logical vector (TRUE for an outlier)

Examples

# Use outliersUV() to detect univariate outliers in an anthropometric
# dataset from a SMART survey from Angola (rl.ex01)
svy <- rl.ex01
svy[outliersUV(svy$muac), ]

Plot helper function for ageChildren() function

Description

Plot helper function for ageChildren() function

Usage

## S3 method for class 'ageChildren'
plot(x, ...)

Arguments

x

Object resulting from applying ageChildren() function
...

Additional barplot() graphical parameters

Value

Bar plot comparing table of observed counts vs table of expected counts

Examples

# Plot Chi-Squared test for age of children in dp.ex02 sample dataset using
# an u5mr of 1 / 10,000 / day.
svy <- dp.ex02
ac <- ageChildren(svy$age, u5mr = 1)
plot(ac)

plot() helper functions for ageHeaping() functions

Description

plot() helper functions for ageHeaping() functions

Usage

## S3 method for class 'ageHeaping'
plot(x, main = "", xlab = "Remainder", ylab = "Frequency", cex = 0.75, ...)

Arguments

x

Object resulting from applying the ageHeaping() function
main

Title of plot
xlab

x-axis label; default is Remainder
ylab

y-axis label; default is Frequency
cex

Character expansion (numeric); default is 0.75
...

Additional plot() graphical parameters

Value

Barplot of frequency of remainders of age when divided by a specified divisor

Examples

# Plot age heaping test results on SMART survey data in Kabul, Afghanistan
# (dp.ex02) using a divisor of 12
svy <- dp.ex02
ah12 <- ageHeaping(svy$age)
plot(ah12)

plot() helper function for digitPreference() function

Description

plot() helper function for digitPreference() function

Usage

## S3 method for class 'digitPreference'
plot(x, main = "", xlab = "Final Digit", ylab = "Frequency", cex = 0.75, ...)

Arguments

x

Object resulting from applying the digitPreference() function.
main

Title of plot
xlab

x-axis label; default is "Final Digit"
ylab

y-axis label; default is "Frequency"
cex

Character expansion; default is 0.75
...

Additional plot() parameters

Value

Plotted output of digitPreference() function comparing the frequencies of the various final digits

Examples

# Plot output of digit preference test applied to anthropometric data from a
# single state from a DHS survey in a West African country
svy <- dp.ex01
digitPreference(svy$wt, digits = 1)
plot(digitPreference(svy$wt, digits = 1))

print() helper function for ageChildren() function

Description

print() helper function for ageChildren() function

Usage

## S3 method for class 'ageChildren'
print(x, ...)

Arguments

x

Object resulting from applying ageChildren() function
...

Additional print() arguments

Value

Printed output of ageChildren() function

Examples

# Print Chi-Squared test for age of children in dp.ex02 sample dataset using
# an u5mr of 1 / 10,000 / day.
svy <- dp.ex02
ac <- ageChildren(svy$age, u5mr = 1)
print(ac)

print() helper functions for ageHeaping() functions

Description

print() helper functions for ageHeaping() functions

Usage

## S3 method for class 'ageHeaping'
print(x, ...)

Arguments

x

Object resulting from applying the ageHeaping() function
...

Additional print() arguments

Value

Printed output of the ageHeaping() function

Examples

# Print age heaping test on SMART survey data in Kabul, Afghanistan (dp.ex02)
# using a divisor of 12
svy <- dp.ex02
ah12 <- ageHeaping(svy$age)
print(ah12)

print() helper function for ageRatioTest() function

Description

print() helper function for ageRatioTest() function

Usage

## S3 method for class 'ageRatioTest'
print(x, ...)

Arguments

x

Object resulting from applying ageRatioTest() function
...

Additional print() arguments

Value

Printed output of ageRatioTest() function

Examples

# Print age-ratio test results for survey dataset from Kabul, Afghanistan
svy <- dp.ex02
print(ageRatioTest(svy$age, ratio = 0.85))

print() helper function for digitPreference() function

Description

print() helper function for digitPreference() function

Usage

## S3 method for class 'digitPreference'
print(x, ...)

Arguments

x

Object resulting from applying the digitPreference() function.
...

Additional print() parameters

Value

Printed output of digitPreference() function

Examples

# Print output of digit preference test applied to anthropometric data from a
#single state from a DHS survey in a West African country
svy <- dp.ex01
print(digitPreference(svy$wt, digits = 1))

print() helper function for print.greensIndex() function

Description

print() helper function for print.greensIndex() function

Usage

## S3 method for class 'greensIndex'
print(x, ...)

Arguments

x

Object resulting from applying the greensIndex() function
...

Additional print() parameters

Value

Printed output of greensIndex() function

Examples

# Apply Green's Index using anthropometric data from a SMART survey in Sudan
# (flag.ex01)
svy <- flag.ex01
svy$flag <- 0
svy$flag <- ifelse(!is.na(svy$haz) & (svy$haz < -6 | svy$haz > 6), svy$flag + 1, svy$flag)
svy$flag <- ifelse(!is.na(svy$whz) & (svy$whz < -5 | svy$whz > 5), svy$flag + 2, svy$flag)
svy$flag <- ifelse(!is.na(svy$waz) & (svy$waz < -6 | svy$waz > 5), svy$flag + 4, svy$flag)
svy <- svy[svy$flag == 0, ]
svy$stunted <- ifelse(svy$haz < -2, 1, 2)
gi <- greensIndex(data = svy, psu = "psu", case = "stunted")
print(gi)

print() helper function for sexRatioTest() function

Description

print() helper function for sexRatioTest() function

Usage

## S3 method for class 'sexRatioTest'
print(x, ...)

Arguments

x

Output resulting from applying the sexRatioTest() function
...

Additional print() parameters

Value

Printed output of sexRatioTest() function

Examples

# Use sexRatioTest() on household roster data from a survey in Tanzania
# (as.ex01) and census data of Tanzania extracted from Wolfram|Alpha knowledge
# engine (as.ex02)
svy <- as.ex01
ref <- as.ex02
censusM <- sum(ref$Males)
censusF <- sum(ref$Females)
srt <- sexRatioTest(svy$sex, codes = c(1, 2), pop = c(censusM, censusF))
print(srt)

print() helper function for skewKurt() function

Description

print() helper function for skewKurt() function

Usage

## S3 method for class 'skewKurt'
print(x, ...)

Arguments

x

Object resulting from applying the skewKurt() function
...

Additional print() parameters

Value

Printed output of skewKurt() function

Examples

# Use skewKurt() on an anthropometric data from a SMART survey in
# Kabul, Afghanistan (dist.ex01)
svy <- dist.ex01
sk <- skewKurt(svy$muac)
print(sk)

Pyramid plot function for creating population pyramids.

Description

Pyramid plot function for creating population pyramids.

Usage

pyramid.plot(
  x,
  g,
  main = paste("Pyramid plot of", deparse(substitute(x)), "by", deparse(substitute(g))),
  xlab = paste(deparse(substitute(g)), "(", levels(as.factor(g))[1], "/",
    levels(as.factor(g))[2], ")"),
  ylab = deparse(substitute(x)),
  col = "white",
  ...
)

Arguments

x

Vector of ages (usually grouped)
g

Vector of groups (usually sex)
main

Plot title
xlab

x-axis label
ylab

y-axis label
col

Colours for bars. Either a single colour (default is col = "white") for all bars, two colours (e.g. col = c("lightblue", "pink")) for left hand side bars and right hand side bars respectively, or many colours allocated on a checkerboard basis to each bar
...

Other graphical parameters

Value

A table of x by g (invisible)

Examples

# Use pyramid.plot() on anthropometric data from a SMART survey in
# Kabul, Afghanistan (dp.ex02)
svy <- dp.ex02
pyramid.plot(svy$age, svy$sex)

Normal quantile-quantile plot

Description

Normal quantile-quantile plot

Usage

qqNormalPlot(x)

Arguments

x

A numeric vector

Examples

# qqNormalPlot() with data from a SMART survey in Kabul, Afghanistan
# (dist.ex01)
svy <- dist.ex01
qqNormalPlot(svy$muac)
qqNormalPlot(svy$haz)
qqNormalPlot(svy$waz)
qqNormalPlot(svy$whz)

Example dataset for checking ranges and legal values

Description

Anthropometric data from a SMART survey in Angola.

Usage

rl.ex01

Format

A data frame with 906 observations and 6 variables

Variable Description
age Age (months)
sex Gender (1 = Male / 2 = Female)
weight Weight (kg)
height Height (cm)
muac Mid-upper arm circumference (mm)
oedema Presence or absence of oedema

Sex Ratio Test

Description

Sex Ratio Test

Usage

sexRatioTest(sex, codes = c(1, 2), pop = c(1, 1))

Arguments

sex

A vector of values that indicate sex
codes

Codes used to identify males and females (in that order)
pop

Relative populations of males and females (in that order)

Value

A list of class "sexRatioTest" with:

Variable Description
pM Observed proportion male
eM Expected proportion male
X2 Chi-squared test statistic
df Degrees of freedom for Chi-squared test
p p-value for Chi-squared test

Examples

# Use sexRatioTest() on household roster data from a survey in Tanzania
# (as.ex01) and census data of Tanzania extracted from Wolfram|Alpha knowledge
# engine (as.ex02)
svy <- as.ex01
ref <- as.ex02
censusM <- sum(ref$Males)
censusF <- sum(ref$Females)
sexRatioTest(svy$sex, codes = c(1, 2), pop = c(censusM, censusF))

Skew and kurtosis

Description

Skew and kurtosis

Usage

skewKurt(x)

Arguments

x

Numeric vector

Value

A list of class "skewKurt" with:

Variable Description
s Skewness with direction
s.se Standard error of skewness
s.z Test statistic (s.z = s / s.se)
s.p p-value (s != 0)
k Excess kurtosis with direction
k.se Standard error of excess kurtosis
k.z Test statistic (k.z = k / k.se)
k.p p-value (k != 0)

Examples

# Use skewKurt() on an anthropometric data from a SMART survey in
# Kabul, Afghanistan (dist.ex01)
svy <- dist.ex01
skewKurt(svy$muac)

Example dataset for using scatterplots to identify outliers

Description

Anthropometric data from a SMART survey in the Democratic Republic of Congo.

Usage

sp.ex01

Format

A data frame with 895 observations and 6 variables

Variable Description
age Age (months)
sex Gender (1 = Male / 2 = Female)
weight Weight (kg)
height Height (cm)
muac Mid-upper arm circumference (mm)
oedema Presence or absence of oedema

Example dataset for using scatterplots to identify outliers

Description

Anthropometric data from a survey of school-age (i.e., between 5 and 15 years) children from Pakistan

Usage

sp.ex02

Format

A data frame with 849 observations and 9 variables

Variable Description
region Region code
school School code
ageMonths Age (months)
sex Sex (1 = Male / 2 = Female)
weight Weight (kg)
height Height (cm)
haz Height-for-age z-score
waz Weight-for-age z-score
baz Body mass index (BMI)-for-age z-score