| Title: | Epidemiological Data Display Package |
|---|---|
| Description: | Package for data exploration and result presentation. Full 'epicalc' package with data management functions is available at '<https://medipe.psu.ac.th/epicalc/>'. |
| Authors: | Virasakdi Chongsuvivatwong <[email protected]> |
| Maintainer: | Virasakdi Chongsuvivatwong <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 3.5.0.2 |
| Built: | 2024-11-09 04:22:08 UTC |
| Source: | https://github.com/cran/epiDisplay |
This dataset contains data on age at first marriage of attendants at a workshop in 1997.
data(Marryage)data(Marryage)
A data frame with 27 observations on the following 7 variables.
ida numeric vector
sexa factor with levels male female
birthyra numeric vector indicating year of birth
educa factor with levels bach- bachelor or higher
maritala factor with levels Single Married
maryra numeric vector indicating year of marriage
endyra numeric vector indicating year of analysis
data(Marryage) des(Marryage)data(Marryage) des(Marryage)
Split the numeric variable into subsets, compute summary statistics for each, and return the results in a data frame.
## S3 method for class 'numeric' aggregate(x, by, FUN=c("count","sum","mean","median","sd","se","min","max"), na.rm=TRUE, length.warning=TRUE, ...)## S3 method for class 'numeric' aggregate(x, by, FUN=c("count","sum","mean","median","sd","se","min","max"), na.rm=TRUE, length.warning=TRUE, ...)
x |
a numeric variable |
by |
a list of grouping elements, each as long as the variables in 'x'. Names for the grouping variables are provided if they are not given. The elements of the list will be coerced to factors (if they are not already factors). |
FUN |
scalar functions to compute the summary statistics which can be applied to all data subsets. |
na.rm |
whether missing values will be removed during the computation of the statistics. |
length.warning |
show warning if x has any missing values |
... |
additional arguments passed on to 'aggregate' |
This is the 'aggregate' method for objects inheriting from class 'numeric'.
If Epicalc is loaded, applying 'aggregate' to a numeric variable 'x' will call 'aggregate.numeric'. If 'x' is a data frame, 'aggregate.data.frame' will be called.
If the Epicalc package is not loaded, 'aggregate', from the stats package, coerces numeric variables (including 'ts' objects) into a data frame and calls 'aggregate.data.frame'.
The 'FUN' argument in 'aggregate.data.frame' can accept only one function.
'aggregate.numeric' takes a different approach. More than one function can be suppplied to the 'FUN' argument, however it can only be applied to one numeric variable.
'aggregate' in Epicalc is 'backward compatible' with the 'aggregate' function from the stats package. In other words, Epicalc users do not need to change basic syntax or arguments. However, the naming system of the returned object is slightly different. In addition to the ability to provide more statistics in one command, another useful feature of 'aggregate.numeric' in Epicalc is the default values of FUN. Without typing such an argument, 'aggregate.numeric' gives commonly wanted statistics in a shorter line of command.
Note that 'na.rm' set to TRUE by default to allow computation of descriptive statistics such as 'mean', and 'sd', when they are in the FUN argument, and 'length' is computed with missing records included. In standard R functions, the equivalent argument is '"na.rm"=TRUE'.
The default value of the argument 'length.warning' is TRUE. A condition where 'x' has any missing value will be noticed, which is useful during data exploration. In further analysis, after missing values have been recognized, users may change 'length.warning' to FALSE to make the output look nicer. Both 'na.rm' and 'length.,warning' will have no effect if there are no missing values in x.
'count' is an additional function specific to 'aggregate.numeric'. It displays the number of non-missing records in each subgroup.
'aggregate.plot' makes use of the above function in drawing bar plots with error lines computed from 'aggregate.numeric'. When 'FUN="mean"', the automactic choice of error values is "se". Users can also choose "sd" or "ci". 'alpha' is effective only for 'error="ci"'. If 'FUN="median"', the error values are inter-quartile range.
Virasakdi Chongsuvivatwong [email protected]
'aggregate', 'summ' and 'tapply'
data(Compaq) .data <- Compaq attach(.data) ## If 'x' is a data frame, the default S3 aggregate method from the stats package is called. aggregate(data.frame(id,year), by=list(HOSPITAL=hospital, STAGE=stage), FUN="mean") # The two additional columns are means of 'id' and 'year' ## If 'x' is a numeric vector, 'aggregate.numeric' from Epicalc package is called. aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN = mean) # The above command is the same as the one below. # However, note the difference in the name of the last column of the returned # data frame. aggregate.data.frame(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN = mean) # aggregate in Epicalc can handle multiple functions aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN = c("mean", "sd", "length")) ## Handling of missing values .data$year[8] <- NA detach(.data) attach(.data) aggregate(year, by = list(STAGE = stage), FUN = c("length", "count")) # Note the difference between 'length' and 'count' in Stage 1 # Means of subsets in 'aggregrate.data.frame' # have 'na.rm' set to FALSE. aggregate.data.frame(year, by = list(STAGE = stage), FUN = "mean") ## The default value of 'na.rm' is TRUE in aggregate.numeric of Epicalc. aggregate(year, by = list(STAGE = stage), FUN = c("mean","median")) ## It can be set to FALSE though. aggregate(year, by = list(STAGE = stage), FUN = c("mean","median"), "na.rm"=FALSE) # Omitting the FUN argument produces various statistics. options(digits=3) aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage)) # Warning of na.rm aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), length.warning=FALSE) # Newly defined functions can be used p05 <- function(x) quantile(x, prob=.05, na.rm=TRUE) p95 <- function(x) quantile(x, prob=.95, na.rm=TRUE) aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN=c("p05", "p95")) detach(.data) rm(list=ls())data(Compaq) .data <- Compaq attach(.data) ## If 'x' is a data frame, the default S3 aggregate method from the stats package is called. aggregate(data.frame(id,year), by=list(HOSPITAL=hospital, STAGE=stage), FUN="mean") # The two additional columns are means of 'id' and 'year' ## If 'x' is a numeric vector, 'aggregate.numeric' from Epicalc package is called. aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN = mean) # The above command is the same as the one below. # However, note the difference in the name of the last column of the returned # data frame. aggregate.data.frame(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN = mean) # aggregate in Epicalc can handle multiple functions aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN = c("mean", "sd", "length")) ## Handling of missing values .data$year[8] <- NA detach(.data) attach(.data) aggregate(year, by = list(STAGE = stage), FUN = c("length", "count")) # Note the difference between 'length' and 'count' in Stage 1 # Means of subsets in 'aggregrate.data.frame' # have 'na.rm' set to FALSE. aggregate.data.frame(year, by = list(STAGE = stage), FUN = "mean") ## The default value of 'na.rm' is TRUE in aggregate.numeric of Epicalc. aggregate(year, by = list(STAGE = stage), FUN = c("mean","median")) ## It can be set to FALSE though. aggregate(year, by = list(STAGE = stage), FUN = c("mean","median"), "na.rm"=FALSE) # Omitting the FUN argument produces various statistics. options(digits=3) aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage)) # Warning of na.rm aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), length.warning=FALSE) # Newly defined functions can be used p05 <- function(x) quantile(x, prob=.05, na.rm=TRUE) p95 <- function(x) quantile(x, prob=.95, na.rm=TRUE) aggregate(year, by = list(HOSPITAL = hospital, STAGE = stage), FUN=c("p05", "p95")) detach(.data) rm(list=ls())
Split a numeric variable into subsets, plot summary statistics for each
## S3 method for class 'plot' aggregate(x, by, grouping = NULL, FUN = c("mean", "median"), error = c("se", "ci", "sd", "none"), alpha = 0.05, lwd = 1, lty = "auto", line.col = "auto", bin.time = 4, bin.method = c("fixed", "quantile"), legend = "auto", legend.site = "topright", legend.bg = "white", xlim = "auto", ylim = "auto", bar.col = "auto", cap.size = 0.02, lagging = 0.007, main = "auto", return.output = FALSE, ...)## S3 method for class 'plot' aggregate(x, by, grouping = NULL, FUN = c("mean", "median"), error = c("se", "ci", "sd", "none"), alpha = 0.05, lwd = 1, lty = "auto", line.col = "auto", bin.time = 4, bin.method = c("fixed", "quantile"), legend = "auto", legend.site = "topright", legend.bg = "white", xlim = "auto", ylim = "auto", bar.col = "auto", cap.size = 0.02, lagging = 0.007, main = "auto", return.output = FALSE, ...)
x |
a numeric variable |
by |
a list of grouping elements for the bar plot, or a single numeric or integer variable which will form the X axis for the time line graph |
grouping |
further stratification variable for the time line graph |
FUN |
either "mean" or "median" |
error |
statistic to use for error lines (either 'se' or 'sd' for barplot, or 'ci' or 'none' for time line graph). When FUN = "median", can only be 'IQR' (default) or 'none'. |
alpha |
level of significance for confidence intervals |
lwd |
relative width of the "time" lines. See 'lwd' in ?par |
lty |
type of the "time" lines. See 'lty' in ?par |
line.col |
colour(s) of the error and time lines |
bin.time |
number bins in the time line graph |
bin.method |
method to allocate the "time" variable into bins, either with 'fixed' interval or equally distributed sample sizes based on quantiles |
legend |
presence of automatic legend for the time line graph |
legend.site |
a single character string indicating location of the legend. See details of ?legend |
legend.bg |
background colour of the legend |
xlim |
X axis limits |
ylim |
Y axis limits |
bar.col |
bar colours |
cap.size |
relative length of terminating cross-line compared to the range of X axis |
lagging |
lagging value of the error bars of two adjecant categories at the same time point. The value is result of dividing this distance with the range of X axis |
main |
main title of the graph |
return.output |
whether the dataframe resulted from aggregate should be returned |
... |
additional graphic parameters passed on to other methods |
This function plots aggregated values of 'x' by a factor (barplot) or a continuous variable (time line graph).
When 'by' is of class 'factor', a bar plot with error bars is displayed.
When 'by' is a continuous variable (typically implying time), a time line graph is displayed.
Both types of plots have error arguments. Choices are 'se' and 'sd' for the bar plot and 'ci' and IQR for both bar plot and time line graph. All these can be suppressed by specifying 'error'="none".
'bin.time' and 'bin.method' are exclusively used when 'by' is a continuous variable and does not have regular values (minimum frequency of 'by' <3). This condition is automatically and silently detected by 'aggregate.plot' before 'bin.method' chooses the method for aggregation and bin.time determines the number of bins.
If 'legend = TRUE" (by default), a legend box is automatically drawn on the "topright" corner of the graph. This character string can be changed to others such as, "topleft", "center", etc (see examples).
'cap.size' can be assigned to zero to remove the error bar cap.
Virasakdi Chongsuvivatwong [email protected]
'aggregate.data.frame', 'aggregate.numeric', 'tapply'
data(Compaq) .data <- Compaq attach(.data) aggregate.plot(x=year, by=list(HOSPITAL = hospital, STAGE = stage), return = TRUE) # moving legend and chaging bar colours aggregate.plot(x=year, by=list(HOSPITAL = hospital, STAGE = stage), error="ci", legend.site = "topleft", bar.col = c("red","blue")) detach(.data) # Example with regular time intervals (all frequencies > 3) data(Sitka, package="MASS") .data <- Sitka attach(.data) tab1(Time, graph=FALSE) # all frequencies > 3 aggregate.plot(x=size, by=Time, cap.size = 0) # Note no cap on error bars # For black and white presentation aggregate.plot(x=size, by=Time, grouping=treat, FUN="median", line.col=3:4, lwd =2) detach(.data) # Example with irregular time intervals (some frequencies < 3) data(BP) .data <- BP attach(.data) des(.data) age <- as.numeric(as.Date("2008-01-01") - birthdate)/365.25 aggregate.plot(x=sbp, by=age, grouping=saltadd, bin.method="quantile") aggregate.plot(x=sbp, by=age, grouping=saltadd, lwd=3, line.col=c("blue","green") , main = NULL) title(main="Effect of age and salt adding on SBP", xlab="years",ylab="mm.Hg") points(age[saltadd=="no"], sbp[saltadd=="no"], col="blue") points(age[saltadd=="yes"], sbp[saltadd=="yes"], pch=18, col="green") detach(.data) rm(list=ls()) ## For a binary outcome variable, aggregrated probabilities is computed data(Outbreak) .data <- Outbreak attach(.data) .data$age[.data$age == 99] <- NA detach(.data) attach(.data) aggregate.plot(diarrhea, by=age, bin.time=5) diarrhea1 <- factor(diarrhea) levels(diarrhea1) <- c("no","yes") aggregate.plot(diarrhea1, by=age, bin.time=5) detach(.data) rm(list=ls())data(Compaq) .data <- Compaq attach(.data) aggregate.plot(x=year, by=list(HOSPITAL = hospital, STAGE = stage), return = TRUE) # moving legend and chaging bar colours aggregate.plot(x=year, by=list(HOSPITAL = hospital, STAGE = stage), error="ci", legend.site = "topleft", bar.col = c("red","blue")) detach(.data) # Example with regular time intervals (all frequencies > 3) data(Sitka, package="MASS") .data <- Sitka attach(.data) tab1(Time, graph=FALSE) # all frequencies > 3 aggregate.plot(x=size, by=Time, cap.size = 0) # Note no cap on error bars # For black and white presentation aggregate.plot(x=size, by=Time, grouping=treat, FUN="median", line.col=3:4, lwd =2) detach(.data) # Example with irregular time intervals (some frequencies < 3) data(BP) .data <- BP attach(.data) des(.data) age <- as.numeric(as.Date("2008-01-01") - birthdate)/365.25 aggregate.plot(x=sbp, by=age, grouping=saltadd, bin.method="quantile") aggregate.plot(x=sbp, by=age, grouping=saltadd, lwd=3, line.col=c("blue","green") , main = NULL) title(main="Effect of age and salt adding on SBP", xlab="years",ylab="mm.Hg") points(age[saltadd=="no"], sbp[saltadd=="no"], col="blue") points(age[saltadd=="yes"], sbp[saltadd=="yes"], pch=18, col="green") detach(.data) rm(list=ls()) ## For a binary outcome variable, aggregrated probabilities is computed data(Outbreak) .data <- Outbreak attach(.data) .data$age[.data$age == 99] <- NA detach(.data) attach(.data) aggregate.plot(diarrhea, by=age, bin.time=5) diarrhea1 <- factor(diarrhea) levels(diarrhea1) <- c("no","yes") aggregate.plot(diarrhea1, by=age, bin.time=5) detach(.data) rm(list=ls())
Deaths in London from 1st-15th Dec 1952
data(SO2)data(SO2)
A data frame with 15 observations on the following 4 variables.
daya numeric vector: the day in Dec 1952
deathsa numeric vector: number of deaths
smokea numeric vector: atmospheric smoke (mg/cu.m)
SO2a numeric vector: atmospheric sulphur dioxide (parts/million)
from John F. Osborn, Statistical Exercises in Medical Research, Blackwell 1979
Calculate reliability coefficient of items in a data frame
alpha (vars, dataFrame, casewise = FALSE, reverse = TRUE, decimal = 4, vars.to.reverse = NULL, var.labels = TRUE, var.labels.trunc =150) alphaBest (vars, dataFrame, standardized = FALSE)alpha (vars, dataFrame, casewise = FALSE, reverse = TRUE, decimal = 4, vars.to.reverse = NULL, var.labels = TRUE, var.labels.trunc =150) alphaBest (vars, dataFrame, standardized = FALSE)
vars |
a vector containing at least three variables from the data frame |
dataFrame |
data frame where items are set as variables |
casewise |
whether only records with complete data will be used |
reverse |
whether item(s) negatively correlated with other majority will be reversed prior to computation |
decimal |
number of decimal places displayed |
var.labels |
presence of descriptions of variables in the last column of the output |
var.labels.trunc |
number of characters used for variable descriptions, long labels can be truncated |
vars.to.reverse |
variable(s) to reverse prior to computation |
standardized |
whether choosing the best subset of items is based on the standardized alpha coefficient, if FALSE then the unstandardized alpha coefficient is used |
This function is based on the 'reliability' function from package 'Rcmdr', which computes Cronbach's alpha for a composite scale.
There must be at least three items in 'vars' specified by their names or their index in the data frame.
The argument 'reverse' (default = TRUE) automatically reverses items negatively correlated with other majority into negative and reports the activities in the first column of the last result section. This can be overwritten by the argument 'vars.to.reverse'
Similar to the 'reliability' function, users can see the effect of removing each item on the coefficents and the item-rest correlation.
'alphaBest' is a variant of 'alpha' for successive removal of items aiming to reach the highest possible Cronbach alpha. The resultant values include variable indices of excluded and remaining items, which can be forwarded to 'tableStack' to achieve total and mean scores of the best selected items. However, there is no promise that this will give the highest possible alpha. Manual attemps may also be useful in making comparison.
A list.
'alpha' returns an object of class "alpha"
alpha |
unstandardized alpha coefficient |
std.alpha |
standardized alpha coefficient |
sample.size |
sample size |
use.method |
method for handling missing values |
rbar |
the average inter-item correlation |
items.selected |
names of variables included in the function |
alpha.if.removed |
a matrix of unstandardized and standardized alpha coefficients and correlation of each item with the rest of the items |
result |
as above but includes a column showing the items that were reversed (if TRUE) and a column of item description. As a matrix, it could be sent to a spreadsheet software using 'write.csv' |
decimal |
decimal places |
item.labels |
a character vector containing descriptions of the items |
'apha.Best' returns a list of the following elements
best.alpha |
the possible highest alpha obtained from the function |
removed |
indices of items removed by the function |
remaining |
indices of the remaining items |
items.reversed |
names of items reversed |
Virasakdi Chongsuvivatwong [email protected]
'cronbach' from 'psy' package and 'reliability' from 'Rcmdr' package and 'tableStack' and 'unclassDataframe' of Epicalc
data(Cars93, package="MASS") .data <- Cars93 attach(.data) alpha(vars=c(Min.Price:MPG.highway, EngineSize), .data) detach(.data) data(Attitudes) .data <-Attitudes attach(.data) alpha(qa1:qa18, .data) # Needs full screen of Rconsole alpha(qa1:qa18, var.labels.trunc=30, .data) # Fits in with default R console screen alpha(qa1:qa18, reverse=FALSE, .data) alphaBest(qa1:qa18, .data) -> best.alpha best.alpha # .7621 tableStack(best.alpha$remaining, dataFrame=.data, reverse=TRUE) # Manual attempts by trial and error give the following alpha(c(qa1:qa9, qa15,qa18), .data) # .7644 detach(.data) rm(list=ls())data(Cars93, package="MASS") .data <- Cars93 attach(.data) alpha(vars=c(Min.Price:MPG.highway, EngineSize), .data) detach(.data) data(Attitudes) .data <-Attitudes attach(.data) alpha(qa1:qa18, .data) # Needs full screen of Rconsole alpha(qa1:qa18, var.labels.trunc=30, .data) # Fits in with default R console screen alpha(qa1:qa18, reverse=FALSE, .data) alphaBest(qa1:qa18, .data) -> best.alpha best.alpha # .7621 tableStack(best.alpha$remaining, dataFrame=.data, reverse=TRUE) # Manual attempts by trial and error give the following alpha(c(qa1:qa9, qa15,qa18), .data) # .7644 detach(.data) rm(list=ls())
This dataset contains frequency of various combinations of methods of antenatal care in two clinics with the outcome being perinatal mortality.
data(ANCtable)data(ANCtable)
A data frame with 8 observations on the following 4 variables.
deatha numeric vector: 1=no, 2=yes
anca numeric vector indicating antenatal care type: 1=old 2=new
clinica numeric vector indicating clinic code: 1=clinic A, 2=clinic B
Freqa numeric vector of frequencies
data(ANCtable) glm1 <- glm(death==2 ~ factor(anc) + factor(clinic),weights=Freq, family=binomial, data=ANCtable) logistic.display(glm1) glm2 <- glm(death==2 ~ factor(anc) + factor(clinic),weights=Freq, family=binomial, data=ANCtable) summary(glm2)$coefficientsdata(ANCtable) glm1 <- glm(death==2 ~ factor(anc) + factor(clinic),weights=Freq, family=binomial, data=ANCtable) logistic.display(glm1) glm2 <- glm(death==2 ~ factor(anc) + factor(clinic),weights=Freq, family=binomial, data=ANCtable) summary(glm2)$coefficients
This dataset contains records of high risk pregnant women under a trial on new and old methods of antenatal care in two clinics. The outcome was perinatal mortality.
data(ANCdata)data(ANCdata)
A data frame with 755 observations on the following 3 variables.
deatha factor with levels no yes
anca factor with levels old new
clinica factor with levels A B
Survey on attitudes related to services among hospital staff.
Codes for the answers qa1 to qa18 are
| 1 | = strongly disagree | |
| 2 | = disagree | |
| 3 | = neutral | |
| 4 | = agree | |
| 5 | = strong agree | |
data(Attitudes)data(Attitudes)
A data frame with 136 observations on the following 7 variables.
ididentifying code of repondent
sexgender of respondent
depcode of department
qa1I have pride in my job
qa2I'm happy to give service
qa3I feel difficulty in giving service
qa4I can improve my service
qa5A service person must have patience
qa6I would change my job if had the chance
qa7Devoting some personal time will improve oneself
qa8Hard work will improve oneself
qa9Smiling leads to trust
qa10I feel bad if I cannot give service
qa11A client is not always right
qa12Experienced clients should follow the procedure
qa13A client violating the regulation should not bargain
qa14Understanding colleagues will lead to understanding clients
qa15Clients like this place due to good service
qa16Clients who expect our smiling faces create pressure on us
qa17Clients are often self-centered
qa18Clients should be better served
The file consists of a subsample of 1934 women grouped in 60 districts.
data(Bang)data(Bang)
A data frame with 1934 observations on the following 7 variables.
womanidentifying code of each woman
districtidentifying code for each district
user1 = using contraceptive 0 = not using
living.childrenNumber of living children at time of survey
| 1 | = none | |
| 2 | = 1 | |
| 3 | = 2 | |
| 4 | = 3 or more | |
age_meanage of woman in years, centred around the mean
urbanType of region of residence: 1 = urban, 0 = rural
constantconstant term = 1
Huq, N. M., and Cleland, J. 1990. Bangladesh Fertility Survey 1989 (Main Report). Dhaka: National Institute of Population Research and Training
This dataset contains information on the records of 100 adults from a small cross-sectional survey in 2001 investigating blood pressure and its determinants in a community.
data(BP)data(BP)
A data frame containing 100 observations and 6 variables with variable descriptions.
data(BP) des(BP)data(BP) des(BP)
A dataset on cancer survival checking whether there is a survival difference between cancer patients in private and public hospitals.
data(Compaq)data(Compaq)
A data frame with 1064 observations on the following 7 variables.
ida numeric vector
hospitala factor with levels Public hospital Private hospital
statusa numeric vector
stagea factor with levels Stage 1 Stage 2 Stage 3 Stage 4
agegra factor with levels <40 40-49 50-59 60+
sesa factor with levels Rich High-middle Poor-middle Poor
yeara numeric vector indicating the year of recruitment into the study
data(Compaq) des(Compaq)data(Compaq) des(Compaq)
Odds ratio calculation and graphing
cc(outcome, exposure, decimal = 2, cctable = NULL, graph = TRUE, original = TRUE, design = "cohort", main, xlab = "auto", ylab, alpha = .05, fisher.or = FALSE, exact.ci.or = FALSE) cci(caseexp, controlex, casenonex, controlnonex, cctable = NULL, graph = TRUE, design = "cohort", main, xlab, ylab, xaxis, yaxis, alpha = .05, fisher.or = FALSE, exact.ci.or = FALSE,decimal = 2 ) cs(outcome, exposure, cctable = NULL, decimal = 2, method="Newcombe.Wilson", main, xlab, ylab, cex, cex.axis) csi(caseexp, controlex, casenonex, controlnonex, cctable = NULL, decimal = 2, method="Newcombe.Wilson") graph.casecontrol(caseexp, controlex, casenonex, controlnonex, decimal=2) graph.prospective(caseexp, controlex, casenonex, controlnonex, decimal=2) labelTable(outcome, exposure, cctable = NULL, cctable.dimnames = NULL) make2x2(caseexp, controlex, casenonex, controlnonex)cc(outcome, exposure, decimal = 2, cctable = NULL, graph = TRUE, original = TRUE, design = "cohort", main, xlab = "auto", ylab, alpha = .05, fisher.or = FALSE, exact.ci.or = FALSE) cci(caseexp, controlex, casenonex, controlnonex, cctable = NULL, graph = TRUE, design = "cohort", main, xlab, ylab, xaxis, yaxis, alpha = .05, fisher.or = FALSE, exact.ci.or = FALSE,decimal = 2 ) cs(outcome, exposure, cctable = NULL, decimal = 2, method="Newcombe.Wilson", main, xlab, ylab, cex, cex.axis) csi(caseexp, controlex, casenonex, controlnonex, cctable = NULL, decimal = 2, method="Newcombe.Wilson") graph.casecontrol(caseexp, controlex, casenonex, controlnonex, decimal=2) graph.prospective(caseexp, controlex, casenonex, controlnonex, decimal=2) labelTable(outcome, exposure, cctable = NULL, cctable.dimnames = NULL) make2x2(caseexp, controlex, casenonex, controlnonex)
cctable.dimnames |
Dimension names of the variables, usually omitted |
decimal |
number of decimal places displayed |
outcome, exposure
|
two dichotomous variables |
cctable |
A 2-by-2 table. If specified, will supercede the outcome and exposure variables |
graph |
If TRUE (default), produces an odds ratio plot |
design |
Specification for graph; can be "case control","case-control", "cohort" or "prospective" |
caseexp |
Number of cases exposed |
controlex |
Number of controls exposed |
casenonex |
Number of cases not exosed |
controlnonex |
Number of controls not exposed |
original |
should the original table be displayed instead of standard outcome vs exposure table |
main |
main title of the graph |
xlab |
label on X axis |
ylab |
label on Y axis |
alpha |
level of significance |
fisher.or |
whether odds ratio should be computed by the exact method |
exact.ci.or |
whether confidence limite of the odds ratio should be computed by the exact method |
xaxis |
two categories of exposure in graph |
yaxis |
two categories of outcome in graph |
method |
method of computation for 95 percent limits of risk difference |
cex.axis |
character expansion factor for graph axis |
cex |
character expansion factor for text in the graph |
'cc' usually reads in two variables whereas in 'cci' four number are entered manually. However, both the variables and the numbers should be omitted if the analysis is directly on a table specified by 'cctable'.
From both functions, odds ratio and its confidence limits, chisquared test and Fisher's exact test are computed. The odds ratio calcuation is based on cross product method unless 'fisher.or' is set as TRUE. It's confidence limits are obtained by the exact method unless exact.ci.or is set as FALSE.
'cs' and 'csi' are for cohort and cross-sectional studies. It computes the absolute risk, risk difference, and risk ratio. When the exposure is a risk factor, the attributable fraction exposure, attributable fraction population and number needed to harm (NNH) are also displayed in the output. When the exposure is a protective factor, protective efficacy or percent of risk reduced and number needed to treat (NNT) are displayed instead.
If there are more than 2 exposure categories and the sample size is large enough, a graph will be plotted.
'method' in 'csi' and 'cs' chooses whether confidence limits of the risk difference should be computed by Newcomb-Wilson method. Both this and the standard method may give non-sensible values if the risk difference is not statistically significant.
'make2x2' creates a 2-by-2 table using the above orientation.
'graph.casecontrol' and 'graph.prospective' draw a graph comparing the odds of exposure between cases and controls or odds of diseased between exposed and non-exposed.
These two graphic commands are automatically chosen by 'cc' and 'cci', depending on the 'design' argument.
Alternatively, a contingency table saved from 'make2x2' can be supplied as the 'cctable' argument for the 'cc' function and so on.
Virasakdi Chongsuvivatwong [email protected]
'fisher.test', 'chisq.test' and 'mhor'
data(Oswego) .data <- Oswego attach(.data) cc(ill, chocolate) cc(ill, chocolate, design="case-control") cs(ill, chocolate) # The outcome variable should come first. # For the following table # chocolate # ill FALSE TRUE # FALSE 7 22 # TRUE 20 25 # cci(25, 22, 20, 7) graph.casecontrol(25, 22, 20, 7) graph.prospective(25, 22, 20, 7) # Each of the above two lines produces untitled graph, which can be decorated # additionally decorated #Alternatively table1 <- make2x2(25,22,20,7) cc(outcome=NULL, exposure=NULL, cctable=table1) cs(outcome=NULL, exposure=NULL, cctable=table1) agegr <- pyramid(age, sex, bin=30)$ageGroup cs(ill, agegr, main="Risk ratio by age group", xlab="Age (years)") rm(list=ls()) detach(.data)data(Oswego) .data <- Oswego attach(.data) cc(ill, chocolate) cc(ill, chocolate, design="case-control") cs(ill, chocolate) # The outcome variable should come first. # For the following table # chocolate # ill FALSE TRUE # FALSE 7 22 # TRUE 20 25 # cci(25, 22, 20, 7) graph.casecontrol(25, 22, 20, 7) graph.prospective(25, 22, 20, 7) # Each of the above two lines produces untitled graph, which can be decorated # additionally decorated #Alternatively table1 <- make2x2(25,22,20,7) cc(outcome=NULL, exposure=NULL, cctable=table1) cs(outcome=NULL, exposure=NULL, cctable=table1) agegr <- pyramid(age, sex, bin=30)$ageGroup cs(ill, agegr, main="Risk ratio by age group", xlab="Age (years)") rm(list=ls()) detach(.data)
Compute confidence interval(s) of variables or values input from keyboard.
ci(x, ...) ## Default S3 method: ci(x,...) ## S3 method for class 'binomial' ci(x, size, precision, alpha = 0.05, ...) ## S3 method for class 'numeric' ci(x, n, sds, alpha = 0.05, ...) ## S3 method for class 'poisson' ci(x, person.time, precision, alpha = 0.05, ...)ci(x, ...) ## Default S3 method: ci(x,...) ## S3 method for class 'binomial' ci(x, size, precision, alpha = 0.05, ...) ## S3 method for class 'numeric' ci(x, n, sds, alpha = 0.05, ...) ## S3 method for class 'poisson' ci(x, person.time, precision, alpha = 0.05, ...)
x |
a variable for 'ci', number of success for 'ci.binomial', mean(s) for 'ci.numeric', and counts for 'ci.poisson' |
size |
denominator for success |
precision |
level of precision used during computation for the confidence limits |
alpha |
significance level |
n |
sample size |
sds |
standard deviation |
person.time |
denominator for count |
... |
further arguments passed to or used by other methods |
These functions compute confidence intervals of probability, mean and incidence from variables in a dataset or values from keyboard input.
'ci' will try to identify the nature of the variable 'x' and determine the appropriate method (between 'ci.binomial' and 'ci.numeric') for computation. 'ci' without a specified method will never call 'ci.poisson'.
The specific method, ie. 'ci.binomial', 'ci.numeric' or 'ci.poisson', should be used when the values are input from the keyboard or from an aggregated data frame with columns of variables for the arguments.
'ci.binomial' and 'ci.numeric' employ exact probability computation while 'ci.numeric' is based on the t-distribution assumption.
'ci.binomial' and 'ci.poisson' return a data frame containing the number of events, the denominator and the incidence rate. 'ci.numeric' returns means and standard deviations. All of these are followed by the standard error and the confidence limit, the level of which is determined by 'alpha'
Virasakdi Chongsuvivatwong [email protected]
'summ'
data(Oswego) .data <- Oswego attach(.data) # logical variable ci(ill) # numeric variable ci(age) # factor ci(sex=="M") ci(sex=="F") detach(.data) # Example of confidence interval for means library(MASS) .data <- Cars93 attach(.data) car.price <- aggregate(Price, by=list(type=Type), FUN=c("mean","length","sd")) car.price ci.numeric(x=car.price$mean, n=car.price$length, sds=car.price$sd.Price ) detach(.data) rm(list=ls()) # Example of confidence interval for probabilty data(ANCdata) .data <- ANCdata attach(.data) death1 <- death=="yes" death.by.group <- aggregate.numeric(death1, by=list(anc=anc, clinic=clinic), FUN=c("sum","length")) death.by.group ci.binomial(death.by.group$sum.death1, death.by.group$length) detach(.data) rm(list=ls()) # Example of confidence interval for incidence data(Montana) .data <- Montana attach(.data) des(.data) age.Montana <- aggregate.data.frame(Montana[,1:2], by=list(agegr=Montana$agegr),FUN="sum") age.Montana ci.poisson(age.Montana$respdeath, person.time=age.Montana$personyrs) detach(.data) rm(list=ls()) # Keyboard input # What is the 95 % CI of sensitivity of a test that gives all # positive results among 40 diseased individuals ci.binomial(40,40) # What is the 99 % CI of incidence of a disease if the number # of cases is 25 among 340,000 person-years ci.poisson(25, 340000, alpha=.01) # 4.1 to 12.0 per 100,000 person-yearsdata(Oswego) .data <- Oswego attach(.data) # logical variable ci(ill) # numeric variable ci(age) # factor ci(sex=="M") ci(sex=="F") detach(.data) # Example of confidence interval for means library(MASS) .data <- Cars93 attach(.data) car.price <- aggregate(Price, by=list(type=Type), FUN=c("mean","length","sd")) car.price ci.numeric(x=car.price$mean, n=car.price$length, sds=car.price$sd.Price ) detach(.data) rm(list=ls()) # Example of confidence interval for probabilty data(ANCdata) .data <- ANCdata attach(.data) death1 <- death=="yes" death.by.group <- aggregate.numeric(death1, by=list(anc=anc, clinic=clinic), FUN=c("sum","length")) death.by.group ci.binomial(death.by.group$sum.death1, death.by.group$length) detach(.data) rm(list=ls()) # Example of confidence interval for incidence data(Montana) .data <- Montana attach(.data) des(.data) age.Montana <- aggregate.data.frame(Montana[,1:2], by=list(agegr=Montana$agegr),FUN="sum") age.Montana ci.poisson(age.Montana$respdeath, person.time=age.Montana$personyrs) detach(.data) rm(list=ls()) # Keyboard input # What is the 95 % CI of sensitivity of a test that gives all # positive results among 40 diseased individuals ci.binomial(40,40) # What is the 99 % CI of incidence of a disease if the number # of cases is 25 among 340,000 person-years ci.poisson(25, 340000, alpha=.01) # 4.1 to 12.0 per 100,000 person-years
Print description, summary statistics and one-way tabulation of variables
codebook(dataFrame)codebook(dataFrame)
dataFrame |
A data frame for printing the codebook |
The default value of dataFrame (ie if no argument is supplied) is '.data'.
While 'summ' produces summary statistics of both numeric and factor variables, 'codebook' gives summary statistics of all numeric variables and one-way tabulation of all factors of the data frame.
Virasakdi Chongsuvivatwong [email protected]
'use', 'summ', 'tab1' and 'tableStack'
data(Familydata) codebook(Familydata)data(Familydata) codebook(Familydata)
The data come from clients of a family planning clinic.
For all variables except id: 9, 99, 99.9, 888, 999 represent missing values
data(Planning)data(Planning)
A data frame with 251 observations on the following 11 variables.
IDa numeric vector: ID code
AGEa numeric vector
RELIGa numeric vector: Religion
| 1 | = Buddhist | |
| 2 | = Muslim | |
PEDa numeric vector: Patient's education level
| 1 | = none | |
| 2 | = primary school | |
| 3 | = secondary school | |
| 4 | = high school | |
| 5 | = vocational school | |
| 6 | = university | |
| 7 | = other | |
INCOMEa numeric vector: Monthly income in Thai Baht
| 1 | = nil | |
| 2 | = < 1,000 | |
| 3 | = 1,000-4,999 | |
| 4 | = 5,000-9,999 | |
| 5 | = 10,000 | |
AMa numeric vector: Age at marriage
REASONa numeric vector: Reason for family planning
| 1 | = birth spacing | |
| 2 | = enough children | |
| 3 | = other | |
BPSa numeric vector: systolic blood pressure
BPDa numeric vector: diastolic blood pressure
WTa numeric vector: weight (Kg)
HTa numeric vector: height (cm)
data(Planning) des(Planning) # Change var. name to lowercase names(Planning) <- tolower(names(Planning)) .data <- Planning des(.data) # Check for duplication of 'id' attach(.data) any(duplicated(id)) duplicated(id) id[duplicated(id)] #215 # Which one(s) are missing? setdiff(min(id):max(id), id) # 216 # Correct the wrong on id[duplicated(id)] <- 216 detach(.data) rm(list=ls())data(Planning) des(Planning) # Change var. name to lowercase names(Planning) <- tolower(names(Planning)) .data <- Planning des(.data) # Check for duplication of 'id' attach(.data) any(duplicated(id)) duplicated(id) id[duplicated(id)] #215 # Which one(s) are missing? setdiff(min(id):max(id), id) # 216 # Correct the wrong on id[duplicated(id)] <- 216 detach(.data) rm(list=ls())
Description of a data frame or a variable or wildcard for variable names
des(dataFrame)des(dataFrame)
dataFrame |
a data frame |
The variable names will be listed with class and the description of each variable
Virasakdi Chongsuvivatwong [email protected]
'summ', 'label.var', 'subset' and 'keepData'
data(Oswego) .data <- Oswego des(.data)data(Oswego) .data <- Oswego des(.data)
Dataset from a community survey on water containers infested by mosquito larvae.
data(DHF99)data(DHF99)
A data frame with 300 observations on the following 5 variables.
houseida numeric vector
villagea numeric vector indicating village ID
educationa factor with levels Primary Secondary High school Bachelor Other
containersa numeric vector indicating number of containers infested
viltypea factor with levels rural urban slum
Thammapalo, S., Chongsuwiwatwong, V., Geater, A., Lim, A., Choomalee, K. 2005. Socio-demographic and environmental factors associated with Aedes breeding places in Phuket, Thailand. Southeast Asian J Trop Med Pub Hlth 36(2): 426-33.
Plot of frequency in dots
dotplot (x, bin = "auto", by = NULL, xmin = NULL, xmax = NULL, time.format = NULL, time.step = NULL, pch = 18, dot.col = "auto", main = "auto", ylab = "auto", cex.X.axis = 1, cex.Y.axis = 1, ...)dotplot (x, bin = "auto", by = NULL, xmin = NULL, xmax = NULL, time.format = NULL, time.step = NULL, pch = 18, dot.col = "auto", main = "auto", ylab = "auto", cex.X.axis = 1, cex.Y.axis = 1, ...)
x |
a numeric vector. Allowed types also include "Date" and "POSIXct" |
bin |
number of bins for the range of 'x' |
by |
stratification variable |
xmin |
lower bound of x in the graph |
xmax |
upper bound of x in the graph |
time.format |
format for time or date at the tick marks |
time.step |
a character string indicating increment of the sequence of tick marks |
pch |
either an integer specifying a symbol or a single character to be used as the default in plotting points |
dot.col |
a character or a numeric vector indicating the colour of each category of 'by' |
main |
main title |
ylab |
Y axis title |
cex.X.axis |
character extension scale of X axis |
cex.Y.axis |
character extension scale of Y axis |
... |
graphical parameters for the dots when there is no stratification |
'dotplot' in Epicalc is similar to a histogram. Each dot represents one record. Attributes of the dots can be further specified in '...' when there is no strafication. Otherwise, the dots are plotted as a diamond shape and the colours are automatically chosen based on the current palette and the number of strata.
When 'bin="auto"' (by default), and the class of the vector is 'integer', 'bin' will be automatically set to max(x)-min(x)+1. This strategy is also applied to all other time and date variables. Users can try other values if the defaults are not to their liking. See the example of 'timeExposed' below.
The argument 'xmin' and 'xmax' indicate the range of x to be displayed on the graph. These two arguments are independent from the value of 'bin', which controls only the number of columns for the original data range.
Dotplot usually starts the first tick mark on the X-axis at 'xmin' (or min(x) if the 'xmin' is not specified). The argument 'time.step' is typically a character string, containing one of 'sec', 'min', 'hour', 'day', 'DSTday', 'week', 'month' or 'year'. This can optionally be preceded by an integer and a space, or followed by "s", such as "2 weeks".
Setting proper 'xmin', 'xmax' and 'time.step' can improve the location of tick marks on the X-axis. The 'time.format' argument can then be given to further improve the graph. See the last two examples for a better understanding.
Virasakdi Chongsuvivatwong [email protected]
'summ', 'hist', 'seq.Date' and 'seq.POSIXt'
a <- rep(1:2, 250) b <- rnorm(500,mean=a) dotplot(b) dotplot(b, pch=1) dotplot(b, by=a) dotplot(b, by=a, pch=1) # You may try other values of 'pch' # For the commands below, # if dates in X axis are not readable, # try omitting '#' from the next line # Sys.setlocale("LC_ALL", "C") # The number of dots in each column is the frequency # of 'x' for the exact value on the X axis. data(Outbreak) .data <- Outbreak attach(.data) class(age) # numeric dotplot(age) # 40 columns age.as.integer <- as.integer(age) dotplot(age.as.integer) # 'bin' is the number of columns in the data range. # Specifying 'min' and 'max' only expands or truncates # the range of the X axis and has no effect on the distribution # of the dots inside the data range. dotplot(age.as.integer, xmin=0, xmax=150) # Just for demonstration. dotplot(age.as.integer, xmin=0, xmax=70) # the "99"s are now out of the plot. dotplot(age.as.integer, xmin=0, xmax=70, by=sex) # Controlling colours of the dots dotplot(age.as.integer, xmin=0, xmax=70, dot.col="chocolate") sex1 <- factor(sex); levels(sex1) <- list("M"=1,"F"=0) dotplot(age.as.integer, xmin=0, xmax=70, by=sex1, dot.col=c(2,5)) dotplot(age.as.integer, xmin=0, xmax=70, by=sex1, dot.col=c("brown","blue"), main="Age by sex", cex.X.axis=1.3, cex.Y.axis=1.5, cex.main=1.5) rm(list=ls()) detach(.data)a <- rep(1:2, 250) b <- rnorm(500,mean=a) dotplot(b) dotplot(b, pch=1) dotplot(b, by=a) dotplot(b, by=a, pch=1) # You may try other values of 'pch' # For the commands below, # if dates in X axis are not readable, # try omitting '#' from the next line # Sys.setlocale("LC_ALL", "C") # The number of dots in each column is the frequency # of 'x' for the exact value on the X axis. data(Outbreak) .data <- Outbreak attach(.data) class(age) # numeric dotplot(age) # 40 columns age.as.integer <- as.integer(age) dotplot(age.as.integer) # 'bin' is the number of columns in the data range. # Specifying 'min' and 'max' only expands or truncates # the range of the X axis and has no effect on the distribution # of the dots inside the data range. dotplot(age.as.integer, xmin=0, xmax=150) # Just for demonstration. dotplot(age.as.integer, xmin=0, xmax=70) # the "99"s are now out of the plot. dotplot(age.as.integer, xmin=0, xmax=70, by=sex) # Controlling colours of the dots dotplot(age.as.integer, xmin=0, xmax=70, dot.col="chocolate") sex1 <- factor(sex); levels(sex1) <- list("M"=1,"F"=0) dotplot(age.as.integer, xmin=0, xmax=70, by=sex1, dot.col=c(2,5)) dotplot(age.as.integer, xmin=0, xmax=70, by=sex1, dot.col=c("brown","blue"), main="Age by sex", cex.X.axis=1.3, cex.Y.axis=1.5, cex.main=1.5) rm(list=ls()) detach(.data)
This case-control study has one case series and two control groups.
The subjects were recruited based on three types of pregnancy outcome
data(Ectopic)data(Ectopic)
A data frame with 723 observations on the following 4 variables.
ida numeric vector
outca factor with levels EP IA Deli
| EP | = ectopic pregnancy | |
| IA | = women coming for induced abortion | |
| Deli | = women admitted for full-term delivery | |
hiaa factor with levels never IA ever IA
gravia factor with levels 1-2 3-4 >4
data(Ectopic) library(nnet) data(Ectopic) .data <- Ectopic multi1 <- multinom(outc ~ hia + gravi, data=.data) summary(multi1) mlogit.display(multi1) # Changing referent group of outcome .data$outcIA <- relevel(.data$outc, ref="IA") multi2 <- multinom(outcIA ~ hia + gravi, data=.data) summary(multi2) mlogit.display(multi2)data(Ectopic) library(nnet) data(Ectopic) .data <- Ectopic multi1 <- multinom(outc ~ hia + gravi, data=.data) summary(multi1) mlogit.display(multi1) # Changing referent group of outcome .data$outcIA <- relevel(.data$outc, ref="IA") multi2 <- multinom(outcIA ~ hia + gravi, data=.data) summary(multi2) mlogit.display(multi2)
Anthropometric and financial data of a hypothetical family
data(Familydata)data(Familydata)
A data frame with 11 observations on the following 6 variables.
codea character vector
agea numeric vector
hta numeric vector
wta numeric vector
moneya numeric vector
sexa factor with levels F M
data(Familydata) .data <- Familydata des(.data) summ(.data) age2 <- with(.data, age)^2 with(.data, plot(age, money, log="y")) dots.of.age <- seq(0,80,0.01) new.data.frame <- data.frame(age=dots.of.age, age2=dots.of.age^2) lm1 <- lm(log(money) ~ age + age2, data=.data) summary(lm1)$coefficients dots.of.money <- predict.lm(lm1, new.data.frame) lines(dots.of.age, exp(dots.of.money), col="blue")data(Familydata) .data <- Familydata des(.data) summ(.data) age2 <- with(.data, age)^2 with(.data, plot(age, money, log="y")) dots.of.age <- seq(0,80,0.01) new.data.frame <- data.frame(age=dots.of.age, age2=dots.of.age^2) lm1 <- lm(log(money) ~ age + age2, data=.data) summary(lm1)$coefficients dots.of.money <- predict.lm(lm1, new.data.frame) lines(dots.of.age, exp(dots.of.money), col="blue")
Plot longitudinal values of individuals with or without stratification
followup.plot(id, time, outcome, by = NULL, n.of.lines = NULL, legend = TRUE, legend.site = "topright", lty = "auto", line.col = "auto", stress = NULL, stress.labels = FALSE, label.col = 1, stress.col = NULL, stress.width = NULL, stress.type = NULL, lwd = 1, xlab, ylab, ...)followup.plot(id, time, outcome, by = NULL, n.of.lines = NULL, legend = TRUE, legend.site = "topright", lty = "auto", line.col = "auto", stress = NULL, stress.labels = FALSE, label.col = 1, stress.col = NULL, stress.width = NULL, stress.type = NULL, lwd = 1, xlab, ylab, ...)
id |
idenfication variable of the same subject being followed up |
time |
time at each measurement |
outcome |
continuous outcome variable |
by |
stratification factor (if any) |
n.of.lines |
number of lines (or number of subjects in the data frame) randomly chosen for drawing |
legend |
whether a legend will be automatically included in the graph |
legend.site |
a single character string indicating location of the legend. See details of ?legend |
lty |
type of the "time" lines. See 'lty' in ?par |
line.col |
line colour(s) for non-stratified plot |
stress |
subset of ids to draw stressed lines |
stress.labels |
whether the stressed lines should be labelled |
label.col |
single integer indicating colour of the stressed line labels |
stress.col |
colour values used for the stressed line. Default value is '1' or black |
stress.width |
relative width of the stressed line |
stress.type |
line type code for the stressed line |
lwd |
line width |
xlab |
label for X axis |
ylab |
label for Y axis |
... |
other graphic parameters |
'followup.plot' plots outcome over time of the individual subjects.
If a stratification variable 'by' is specified, the levels of this variable will be used to color the lines.
'n.of.lines' is used to reduce the number of lines to allow the pattern to be seen more clearly.
'legend' is omitted if 'n.of.lines' is not NULL or the number of subjects exceeds 7 without stratification.
'line.col' works only for a non-stratified plot. It can be a single standard colour or "multicolor".
Values for 'stress.col', 'stress.width' and 'stress.type', if not NULL, should follow those for 'col', 'lwd' and 'lty', respectively
Virasakdi Chongsuvivatwong [email protected]
'plot','lines'
.data <- Indometh attach(.data) followup.plot(Subject, time, conc) followup.plot(Subject, time, conc, lty=1:6, line.col=rep("black",6)) detach(.data) .data <- Sitka attach(.data) followup.plot(tree, Time, size) followup.plot(tree, Time, size, line.col = "brown") followup.plot(tree, Time, size, line.col = "multicolor") followup.plot(tree, Time, size, n.of.lines=20, line.col = "multicolor") # Breakdown of color by treatment group followup.plot(tree, Time, size, by=treat) # The number of lines reduced to 40 followup.plot(tree, Time, size, by=treat, n.of.lines=40) # Stress some lines length(table(tree)) # 79 trees followed up # Identifying trees that sometimes became smaller .data <- .data[order(.data$tree, .data$Time),] detach(.data) attach(.data) next.tree <- c(tree[-1], NA) next.size <- c(size[-1], NA) next.size[tree != next.tree] <- NA smaller.trees <- tree[next.size < size] followup.plot (tree, Time, size, line.col=5, stress=smaller.trees, stress.col=2, stress.width=2, stress.type=2) followup.plot (tree, Time, size, line.col=5, stress=smaller.trees, stress.col=2, stress.width=2, stress.type=2, stress.labels=TRUE) detach(.data) rm(list=ls()).data <- Indometh attach(.data) followup.plot(Subject, time, conc) followup.plot(Subject, time, conc, lty=1:6, line.col=rep("black",6)) detach(.data) .data <- Sitka attach(.data) followup.plot(tree, Time, size) followup.plot(tree, Time, size, line.col = "brown") followup.plot(tree, Time, size, line.col = "multicolor") followup.plot(tree, Time, size, n.of.lines=20, line.col = "multicolor") # Breakdown of color by treatment group followup.plot(tree, Time, size, by=treat) # The number of lines reduced to 40 followup.plot(tree, Time, size, by=treat, n.of.lines=40) # Stress some lines length(table(tree)) # 79 trees followed up # Identifying trees that sometimes became smaller .data <- .data[order(.data$tree, .data$Time),] detach(.data) attach(.data) next.tree <- c(tree[-1], NA) next.size <- c(size[-1], NA) next.size[tree != next.tree] <- NA smaller.trees <- tree[next.size < size] followup.plot (tree, Time, size, line.col=5, stress=smaller.trees, stress.col=2, stress.width=2, stress.type=2) followup.plot (tree, Time, size, line.col=5, stress=smaller.trees, stress.col=2, stress.width=2, stress.type=2, stress.labels=TRUE) detach(.data) rm(list=ls())
Subset of a dataset from an intervention trial of education on personnel and the effect on neonatal mortality. Non-fatal records were randomly selected from the original dataset, just for practice and interpretation of interaction term.
data(Hakimi)data(Hakimi)
A data frame containing 456 observations and 4 variables.
deadneonatal death: 1=yes, 0=no
treatmentintervention programme: 1=yes, 2=no
malpresmalpresentation of fetus: 1=yes, 0=no
birthwtbirth weight for foetus in gram
data(Hakimi) .data <- Hakimi attach(.data) cc(dead, treatment) mhor(dead, treatment, malpres) detach(.data)data(Hakimi) .data <- Hakimi attach(.data) cc(dead, treatment) mhor(dead, treatment, malpres) detach(.data)
A dataset from a cross-sectional survey in 1993 examining hookworm infection
data(HW93)data(HW93)
A data frame with 637 observations on the following 6 variables.
ida numeric vector for personal identification number
epga numeric vector for eggs per gram of faeces
agea numeric vector for age in years
shoea factor for shoe wearing with levels no yes
intensea factor for intensity of infection in epg. with levels 0 1-1,999 2,000+
agegra factor for age group with levels <15 yrs 15-59 yrs 60+ yrs
data(HW93) des(HW93) .data <- HW93 .data$order.intense <- ordered(.data$intense) ord.hw <- polr(ordered(intense) ~ agegr + shoe, data=.data) summary(ord.hw) ordinal.or.display(ord.hw)data(HW93) des(HW93) .data <- HW93 .data$order.intense <- ordered(.data$intense) ord.hw <- polr(ordered(intense) ~ agegr + shoe, data=.data) summary(ord.hw) ordinal.or.display(ord.hw)
A study using radio-isotope to examine daily blood loss and number of hookworms infecting the patients.
data(Suwit)data(Suwit)
A data frame with 15 observations on the following 3 variables.
ida numeric vector
worma numeric vector: number of worms
blossa numeric vector: estimated daily blood loss (mg/day)
Areekul, S., Devakul, K., Viravan, C., Harinasuta, C. 1970 Studies on blood loss, iron absorption and iron reabsorption in hookworm patients in Thailand. Southeast Asian J Trop Med Pub Hlth 1(4): 519-523.
~~ possibly secondary sources and usages ~~
data(Suwit) with(Suwit, plot(worm, bloss, type="n")) with(Suwit, text(worm, bloss, labels=id)) abline(lm(bloss ~ worm, data=Suwit), col="red")data(Suwit) with(Suwit, plot(worm, bloss, type="n")) with(Suwit, text(worm, bloss, labels=id)) abline(lm(bloss ~ worm, data=Suwit), col="red")
This dataset is a subset of WHO IUD trial. It should be merged with IudFollowup and IudDiscontinue
data(IudAdmit)data(IudAdmit)
A data frame containing 918 observations and 4 variables.
ida numeric vector for personal identification number
idatedate of IUD insertion
lmptimetime since last menstrual period
a122type of IUD
data(IudAdmit)data(IudAdmit)
This dataset is a subset of WHO IUD trial. It should be merged with IudAdmit and IudFollowup
data(IudDiscontinue)data(IudDiscontinue)
A data frame containing 398 observations and 3 variables.
ida numeric vector for personal identification number
discdatedate of discontinuation
d23primary reason for discontinuation
data(IudDiscontinue)data(IudDiscontinue)
This dataset is a subset of WHO IUD trial. It should be merged with IudAdmit and IudDiscontinue
data(IudFollowup)data(IudFollowup)
A data frame containing 4235 observations and 6 variables.
ida numeric vector for personal identification number
vlmpdatedate of last mentrual period before this visit
vdatedate of visit
f22lactating
f51IUD threads visible
f61subject continuing
data(IudFollowup)data(IudFollowup)
Measurement of agreement in categorization by 2 or more raters
kap(x, ...) ## Default S3 method: kap(x, ...) ## S3 method for class 'table' kap(x, decimal =3, wttable = NULL, print.wttable = FALSE, ...) ## S3 method for class '2.raters' kap(x, rater2, decimal =3, ...) ## S3 method for class 'm.raters' kap(x, decimal =3, ...) ## S3 method for class 'ByCategory' kap(x, decimal =3, ...)kap(x, ...) ## Default S3 method: kap(x, ...) ## S3 method for class 'table' kap(x, decimal =3, wttable = NULL, print.wttable = FALSE, ...) ## S3 method for class '2.raters' kap(x, rater2, decimal =3, ...) ## S3 method for class 'm.raters' kap(x, decimal =3, ...) ## S3 method for class 'ByCategory' kap(x, decimal =3, ...)
x |
an object serving the first argument for different methods
|
||||||||||||||||
decimal |
number of decimal in the print |
||||||||||||||||
wttable |
cross tabulation of weights of agreement among categories. It can be NULL, "w" and "w2". Applicable only for 'kap.table' and 'kap.2.raters' |
||||||||||||||||
print.wttable |
whether the weights table will be printed out |
||||||||||||||||
rater2 |
a vector or factor containing opinions of the second rater among two raters. |
||||||||||||||||
... |
further arguments passed to or used by other methods. |
There are two different principles for the calculation of the kappa statistic. 'kap.table' and 'kap.2.raters' use two fixed raters whereas 'kap.m.raters' and 'kap.ByCategory' are based on frequency of category of rating an individual received without a requirement that the raters must be fixed.
'kap.table' analyses kappa statistics from a predefined table of agreement of two raters.
'wttable' is important only if the rating can be more than 2 levels. If this argument is left as default or 'NULL', full agreement will be weighted as 1. Partial agreement is considered as non-agreement and weighted as 0.
When 'wttable = "w"' the weights are given by
where i and j index the rows and columns of the ratings and k is the maximum number of possible ratings. A weight of 1 indicates an observation of perfect agreement.
When 'wttable = "w2", the weights are given by
In this case, weights of partial agreements will further increase.
'wttable' can otherwise be defined by the user.
'kap.2.raters' takes two vectors or factors, one for each of the two raters. Cross-tabulation of the two raters is displayed and automatically forwarded for computation of kappa statistic by 'kap.table'.
'kap.m.raters' is used for more than 2 raters. Although the variables are arranged based on columns of individual raters, only the frequency in each category rating is used. This function calculates the frequencies without any display and automatically forwards the results for computation by 'kap.ByCategory'.
'kap.ByCategory' is for the grouped data format, where each category (column) contains the counts for each individual subject being rated. As mentioned above, the frequencies can come from different sets of raters.
Virasakdi Chongsuvivatwong [email protected]
'table'
## Computation of kappa from a table class <- c("Normal","Benign","Suspect","Cancer") raterA <- gl(4,4, label=class) raterB <- gl(4,1,16, label=class) freq <- c(50,2,0,1,2,30,4,3,0,0,20,1,1,3,4,25) table1 <- xtabs(freq ~ raterA + raterB) table1 kap(table1) wt <-c(1,.5,0,0,.5,1,0,0,0,0,1,.8,0,0,.8,1) wttable <- xtabs(wt ~ raterA + raterB) wttable # Agreement between benign vs normal is .5, suspect vs cancer is .8 kap(table1, wttable=wttable, print.wttable=TRUE) # The following two lines are computational possible but inappropriate kap(table1, wttable = "w", print.wttable=TRUE) kap(table1, wttable = "w2", print.wttable=TRUE) ## A data set from 5 raters with 3 possible categories. category.lab <- c("yes","no","Don't know") rater1 <- factor(c(1,1,3,1,1,1,1,2,1,1), labels=category.lab) rater2 <- factor(c(2,1,3,1,1,2,1,2,3,1), labels=category.lab) rater3 <- factor(c(2,3,3,1,1,2,1,2,3,1), labels=category.lab) rater4 <- factor(c(2,3,3,1,3,2,1,2,3,3), labels=category.lab) rater5 <- factor(c(2,3,3,3,3,2,1,3,3,3), labels=category.lab) kap.m.raters(data.frame(rater1,rater2,rater3,rater4,rater5)) # The above is the same as YES <- c(1,2,0,4,3,1,5,0,1,3) NO <- c(4,0,0,0,0,4,0,4,0,0) DONTKNOW <- c(0,3,5,1,2,0,0,1,4,2) kap.ByCategory(data.frame(YES,NO,DONTKNOW)) # Using 'kap.m.raters' for 2 raters is inappropriate. Kappa obtained # from this method assumes that the agreement can come from any two raters, # which is usually not the case. kap.m.raters(data.frame(rater1, rater2)) # 'kap.2.raters' gives correct results kap.2.raters(rater1, rater2) # When there are missing values, rater3[9] <- NA; rater4[c(1,9)] <- NA kap.m.raters(data.frame(rater1,rater2,rater3,rater4,rater5)) # standard errors and other related statistics are not available. # Two exclusive rating categories give only one common set of results. # The standard error is obtainable even if the numbers of raters vary # among individual subjects being rated. totalRaters <- c(2,2,3,4,3,4,3,5,2,4,5,3,4,4,2,2,3,2,4,5,3,4,3,3,2) pos <- c(2,0,2,3,3,1,0,0,0,4,5,3,4,3,0,2,1,1,1,4,2,0,0,3,2) neg <- totalRaters - pos kap.ByCategory(data.frame(neg, pos))## Computation of kappa from a table class <- c("Normal","Benign","Suspect","Cancer") raterA <- gl(4,4, label=class) raterB <- gl(4,1,16, label=class) freq <- c(50,2,0,1,2,30,4,3,0,0,20,1,1,3,4,25) table1 <- xtabs(freq ~ raterA + raterB) table1 kap(table1) wt <-c(1,.5,0,0,.5,1,0,0,0,0,1,.8,0,0,.8,1) wttable <- xtabs(wt ~ raterA + raterB) wttable # Agreement between benign vs normal is .5, suspect vs cancer is .8 kap(table1, wttable=wttable, print.wttable=TRUE) # The following two lines are computational possible but inappropriate kap(table1, wttable = "w", print.wttable=TRUE) kap(table1, wttable = "w2", print.wttable=TRUE) ## A data set from 5 raters with 3 possible categories. category.lab <- c("yes","no","Don't know") rater1 <- factor(c(1,1,3,1,1,1,1,2,1,1), labels=category.lab) rater2 <- factor(c(2,1,3,1,1,2,1,2,3,1), labels=category.lab) rater3 <- factor(c(2,3,3,1,1,2,1,2,3,1), labels=category.lab) rater4 <- factor(c(2,3,3,1,3,2,1,2,3,3), labels=category.lab) rater5 <- factor(c(2,3,3,3,3,2,1,3,3,3), labels=category.lab) kap.m.raters(data.frame(rater1,rater2,rater3,rater4,rater5)) # The above is the same as YES <- c(1,2,0,4,3,1,5,0,1,3) NO <- c(4,0,0,0,0,4,0,4,0,0) DONTKNOW <- c(0,3,5,1,2,0,0,1,4,2) kap.ByCategory(data.frame(YES,NO,DONTKNOW)) # Using 'kap.m.raters' for 2 raters is inappropriate. Kappa obtained # from this method assumes that the agreement can come from any two raters, # which is usually not the case. kap.m.raters(data.frame(rater1, rater2)) # 'kap.2.raters' gives correct results kap.2.raters(rater1, rater2) # When there are missing values, rater3[9] <- NA; rater4[c(1,9)] <- NA kap.m.raters(data.frame(rater1,rater2,rater3,rater4,rater5)) # standard errors and other related statistics are not available. # Two exclusive rating categories give only one common set of results. # The standard error is obtainable even if the numbers of raters vary # among individual subjects being rated. totalRaters <- c(2,2,3,4,3,4,3,5,2,4,5,3,4,4,2,2,3,2,4,5,3,4,3,3,2) pos <- c(2,0,2,3,3,1,0,0,0,4,5,3,4,3,0,2,1,1,1,4,2,0,0,3,2) neg <- totalRaters - pos kap.ByCategory(data.frame(neg, pos))
List all objects visible in the global environment except user created functions.
lsNoFunction()lsNoFunction()
Compared to standard 'ls()', this function displays only the subset of 'ls()' which are not functions.
The member of this list can be removed by 'zap()' but not the set of the functions created.
Virasakdi Chongsuvivatwong [email protected]
'use', 'detach', 'ls', 'rm'
object1 <- 1:5 object2 <- list(a=3, b=5) function1 <- function(x) {x^3 +1} ls() lsNoFunction() ## To show only functions as.character(lsf.str()[])object1 <- 1:5 object2 <- list(a=3, b=5) function1 <- function(x) {x^3 +1} ls() lsNoFunction() ## To show only functions as.character(lsf.str()[])
Systematic replacement of several values of a variable using an array
lookup(x, lookup.array)lookup(x, lookup.array)
x |
a variable |
lookup.array |
a n-by-2 array used for looking up the recoding scheme |
This command is used for changing more than one value of a variable using a n-by-2 look-up array. The first column of the look-up array (index column) must be unique.
If either the variable or the look-up table is character, the result vector will be character.
For changing the levels of a factor variable, 'recode(vars, "old level", "new level")' or 'levels(var) <- ' instead.
Virasakdi Chongsuvivatwong [email protected]
'replace', 'recode'
a <- c( 1, 2, 2, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 5, NA) tx <- rbind(c(1,2),c(2,1),c(3,4),c(4,NA),c(NA,3)) # Swapping values of 1 and 2; rotating 3, 4 and NA new.a <- lookup(a, tx) data.frame(a, new.a) tableA <- table(a, new.a, exclude=NULL) # All non-diagonal cells which are non-zero are the recoded cells. print(tableA, zero=".") ## Character look-up table b <- c(rep(letters[1:4],2), ".", NA) tx1 <- cbind(c(letters[1:5], ".", NA), c("Disease A","Disease B","Disease C", "Disease D","Disease E", NA, "Unknown")) DiseaseName <- lookup(b, tx1) data.frame(b, DiseaseName)a <- c( 1, 2, 2, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 5, NA) tx <- rbind(c(1,2),c(2,1),c(3,4),c(4,NA),c(NA,3)) # Swapping values of 1 and 2; rotating 3, 4 and NA new.a <- lookup(a, tx) data.frame(a, new.a) tableA <- table(a, new.a, exclude=NULL) # All non-diagonal cells which are non-zero are the recoded cells. print(tableA, zero=".") ## Character look-up table b <- c(rep(letters[1:4],2), ".", NA) tx1 <- cbind(c(letters[1:5], ".", NA), c("Disease A","Disease B","Disease C", "Disease D","Disease E", NA, "Unknown")) DiseaseName <- lookup(b, tx1) data.frame(b, DiseaseName)
Likelihood ratio test for objects of class 'glm'
lrtest (model1, model2)lrtest (model1, model2)
model1, model2
|
Two models of class "glm" having the same set of records and the same type ('family' and 'link') |
Likelihood ratio test checks the difference between -2*logLikelihood of the two models against the change in degrees of freedom using a chi-squared test. It is best applied to a model from 'glm' to test the effect of a factor with more than two levels. The records used in the dataset for both models MUST be the same. The function can also be used with "clogit", which does not have real logLikelihood.
Virasakdi Chongsuvivatwong [email protected]
'glm', 'logLik', 'deviance'
model0 <- glm(case ~ induced + spontaneous, family=binomial, data=infert) model1 <- glm(case ~ induced, family=binomial, data=infert) lrtest (model0, model1) lrtest (model1, model0) # same result lrtest (model1, model0) -> a amodel0 <- glm(case ~ induced + spontaneous, family=binomial, data=infert) model1 <- glm(case ~ induced, family=binomial, data=infert) lrtest (model0, model1) lrtest (model1, model0) # same result lrtest (model1, model0) -> a a
Two different datasets for the same matched case-control study. VC1to6 has 1 case : varying number of controls (from 1 to 6) whereas VC1to1 has the number of control reduced to 1 for each case.
data(VC1to1) data(VC1to6)data(VC1to1) data(VC1to6)
A data frame with the following 5 variables.
matseta numeric vector indicating matched set number from 1 to 26
casea numeric vector: 1=case, 0=control
smokinga numeric vector: 1=smoker, 0=non-smoker
rubbera numeric vector: 1=exposed, 0=never exposed to rubber industry
alcohola numeric vector: 1=drinker, 0=non-drinker
Chongsuvivatwong, V. 1990 A case-control study of esophageal cancer in Southern Thailand. J Gastro Hep 5:391–394.
'infert' in the datasets package.
data(VC1to6) .data <- VC1to6 des(.data) with(.data, matchTab(case, alcohol, matset)) rm(.data)data(VC1to6) .data <- VC1to6 des(.data) with(.data, matchTab(case, alcohol, matset)) rm(.data)
Tabulation of outcome vs exposure from a matched case control study
matchTab (case, exposed, strata, decimal)matchTab (case, exposed, strata, decimal)
case |
Outcome variables where 0 = control and 1 = case |
exposed |
Exposure variable where 0 = non-exposed and 1 = exposed |
strata |
Identification number for each matched set |
decimal |
Number of digits displayed after the decimal point |
Tabulation for an unmatched case control study is based on individual records classified by outcome and exposure variables.
Matched tabulation is tallying based on each matched set. The simplest form is McNemar's table where only one case is matched with one control. 'matchTab' can handle 1:m matching where m can vary from 1 to m. A MLE method is then used to compute the conditional odds ratio.
Virasakdi Chongsuvivatwong [email protected]
'table', 'cc' and 'clogit'
.data <- infert ## Not run: # matchTab(case, induced, stratum) # Tabulation successful but OR not computed # because 'induced' is not binary ## End(Not run) attach(.data) ia <- induced > 0 # any induced abortion matchTab(case, ia, stratum) # See also clogit(case ~ ia + strata(stratum), data=infert) detach(.data) rm(list=ls()).data <- infert ## Not run: # matchTab(case, induced, stratum) # Tabulation successful but OR not computed # because 'induced' is not binary ## End(Not run) attach(.data) ia <- induced > 0 # any induced abortion matchTab(case, ia, stratum) # See also clogit(case ~ ia + strata(stratum), data=infert) detach(.data) rm(list=ls())
Mantel-Haenszel odds ratio calculation and graphing from a stratified case-control study
mhor(..., mhtable = NULL, decimal=2, graph = TRUE, design = "cohort")mhor(..., mhtable = NULL, decimal=2, graph = TRUE, design = "cohort")
... |
Three variables viz. 'outcome', 'exposure' and 'stratification'. |
mhtable |
a 2-by-2-by-s table, where s (strata) is more than one |
decimal |
number of decimal places displayed |
graph |
If TRUE (default), produces an odds ratio plot |
design |
Specification for graph; can be "case control","case-control", "cohort" or "prospective" |
'mhor' computes stratum-specific odds ratios and 95 percent confidence intervals and the Mantel-Haenszel odds ratio and chi-squared test is given as well as the homogeneity test. A stratified odds ratio graph is displayed.
Virasakdi Chongsuvivatwong [email protected]
'fisher.test', 'chisq.test'
data(Oswego) with(Oswego, cc(ill, chocolate)) with(Oswego, mhor(ill, chocolate, sex)) mht1 <- with(Oswego, table(ill, chocolate, sex)) dim(mht1) mhor(mhtable=mht1) # same resultsdata(Oswego) with(Oswego, cc(ill, chocolate)) with(Oswego, mhor(ill, chocolate, sex)) mht1 <- with(Oswego, table(ill, chocolate, sex)) dim(mht1) mhor(mhtable=mht1) # same results
Dataset from a cohort study of exposure to arsenic from industry and deaths from respiratory diseases.
data(Montana)data(Montana)
A data frame with 114 observations on the following 6 variables.
respdeatha numeric vector indicating number of deaths from respiratory diseases
personyrsa numeric vector indicating person-years of exposure
agegra numeric vector: 1=40-49, 2=50-59, 3=60-69, 4=70-79)
perioda numeric vector: 1=1938-1949, 2=1950-1959, 3=1960-1969, 4=1970-1977
startinga numeric vector indicating starting period: 1=pre-1925, 2=1925 & after
arsenica numeric vector indicating years of exposure: 1=<1 year, 2=1-4 years, 3=5-14 years, 4=15+ years
This dataset contains information on the records of 75 persons under investigation for the cause of acute food poisoning after a dinner party.
data(Oswego)data(Oswego)
A data frame containing 75 observations and 20 variables.
EpiInfo package
data(Oswego) .data <- Oswego attach(.data) pyramid(age, sex) detach(.data)data(Oswego) .data <- Oswego attach(.data) pyramid(age, sex) detach(.data)
This dataset contains information from an outbreak investigation concerning food poisoning on a sportsday in Thailand 1990.
Dichotomous variables for exposures and symptoms were coded as follow:
| 0 | = no | |
| 1 | = yes | |
| 9 | = missing or unknown | |
data(Outbreak)data(Outbreak)
A data frame with 1094 observations on the following 13 variables.
ida numeric vector
sexa numeric vector
| 0 | = female | |
| 1 | = male | |
agea numeric vector: age in years
| 99 | = missing | |
exptimean AsIs or character vector of exposure times
beefcurrya numeric vector: whether the subject had eaten beefcurry
saltegga numeric vector: whether the subject had eaten salted eggs
eclaira numeric vector: pieces of eclair eaten
| 80 | = ate but could not remember how much | |
| 90 | = totally missing information | |
watera numeric vector: whether the subject had drunk water
onsetan AsIs or character vector of onset times
nauseaa numeric vector
vomitinga numeric vector
abdpaina numeric vector: abdominal pain
diarrheaa numeric vector
Thaikruea, L., Pataraarechachai, J., Savanpunyalert, P., Naluponjiragul, U. 1995 An unusual outbreak of food poisoning. Southeast Asian J Trop Med Public Health 26(1):78-85.
data(Outbreak) .data <- Outbreak # Distribution of reported pieces of eclair taken attach(.data) tab1(eclair) # Defining missing value .data$eclair[.data$eclair > 20] <- NA detach(.data) attach(.data) pieces.of.eclair <- cut(eclair, c(0,1,2,20), include.lowest=TRUE, right=FALSE) tabpct(pieces.of.eclair, diarrhea) rm(list=ls()) detach(.data)data(Outbreak) .data <- Outbreak # Distribution of reported pieces of eclair taken attach(.data) tab1(eclair) # Defining missing value .data$eclair[.data$eclair > 20] <- NA detach(.data) attach(.data) pieces.of.eclair <- cut(eclair, c(0,1,2,20), include.lowest=TRUE, right=FALSE) tabpct(pieces.of.eclair, diarrhea) rm(list=ls()) detach(.data)
Poisson and negative binomial regression are used for modeling count data. This command tests the deviance against the degrees of freedom in the model thus determining whether there is overdispersion.
poisgof(model)poisgof(model)
model |
A Poisson or negative binomial model |
To test the significance of overdispersion of the errors of a Poisson or negative binomial model, the deviance is tested against degrees of freedom using chi-squared distribution. A low P value indicates significant overdispersion.
Virasakdi Chongsuvivatwong [email protected]
‘glm’
library(MASS) quine.pois <- glm(Days ~ Sex/(Age + Eth*Lrn), data = quine, family=poisson) poisgof(quine.pois) quine.nb1 <- glm.nb(Days ~ Sex/(Age + Eth*Lrn), data = quine) poisgof(quine.nb1)library(MASS) quine.pois <- glm(Days ~ Sex/(Age + Eth*Lrn), data = quine, family=poisson) poisgof(quine.pois) quine.nb1 <- glm.nb(Days ~ Sex/(Age + Eth*Lrn), data = quine) poisgof(quine.nb1)
Calculation of power given the results from a study
power.for.2p(p1, p2, n1, n2, alpha = 0.05) power.for.2means(mu1, mu2, n1, n2, sd1, sd2, alpha = 0.05)power.for.2p(p1, p2, n1, n2, alpha = 0.05) power.for.2means(mu1, mu2, n1, n2, sd1, sd2, alpha = 0.05)
p1, p2
|
probabilities of the two samples |
n1, n2
|
sample sizes of the two samples |
alpha |
significance level |
mu1, mu2
|
means of the two samples |
sd1, sd2
|
standard deviations of the two samples |
These two functions compute the power of a study from the given arguments
Virasakdi Chongsuvivatwong [email protected]
'n.for.2means', 'n.for.2p'
# Suppose, in the example found in 'help(n.for.2p)', # given the two proportions are .8 and .6 and the sample size # for each group is 60. power.for.2p(p1=.8, p2=.6, n1=60, n2=60) # 59 percent # If the means of a continuous outcome variable in the same # two groups were 50 and 60 units and the standard deviations were 30 # and 35 units, then the power to detect a statistical significance # would be power.for.2means(mu1=50, mu2=60, sd1=30, sd2=35, n1=60, n2=60) # 39 percent. Note the graphic display# Suppose, in the example found in 'help(n.for.2p)', # given the two proportions are .8 and .6 and the sample size # for each group is 60. power.for.2p(p1=.8, p2=.6, n1=60, n2=60) # 59 percent # If the means of a continuous outcome variable in the same # two groups were 50 and 60 units and the standard deviations were 30 # and 35 units, then the power to detect a statistical significance # would be power.for.2means(mu1=50, mu2=60, sd1=30, sd2=35, n1=60, n2=60) # 39 percent. Note the graphic display
Print results related to Cronbach's alpha
## S3 method for class 'alpha' print(x, ...)## S3 method for class 'alpha' print(x, ...)
x |
object of class 'alpha' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'tableStack'
data(Attitudes) alpha(qa1:qa18, dataFrame=Attitudes) -> a print(a) adata(Attitudes) alpha(qa1:qa18, dataFrame=Attitudes) -> a print(a) a
Print results for cci and cc commands
## S3 method for class 'cci' print(x, ...)## S3 method for class 'cci' print(x, ...)
x |
object of class 'cci' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'cci'
Print description of data frame of a variable
## S3 method for class 'des' print(x, ...)## S3 method for class 'des' print(x, ...)
x |
object of class 'des' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'des'
Print results for kap.Bycategory commands
## S3 method for class 'kap.ByCategory' print(x, ...)## S3 method for class 'kap.ByCategory' print(x, ...)
x |
object of class 'kap.ByCategory' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'kap.ByCategory'
Print results for kap.table commands
## S3 method for class 'kap.table' print(x, ...)## S3 method for class 'kap.table' print(x, ...)
x |
object of class 'kap.table' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'kap.table'
Print results for likelihood ratio test
## S3 method for class 'lrtest' print(x, ...)## S3 method for class 'lrtest' print(x, ...)
x |
object of class 'lrtest' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'logistic.display'
model0 <- glm(case ~ induced + spontaneous, family=binomial, data=infert) model1 <- glm(case ~ induced, family=binomial, data=infert) lrtest (model0, model1) lrtest (model1, model0) -> a amodel0 <- glm(case ~ induced + spontaneous, family=binomial, data=infert) model1 <- glm(case ~ induced, family=binomial, data=infert) lrtest (model0, model1) lrtest (model1, model0) -> a a
Print matched tabulation results
## S3 method for class 'matchTab' print(x, ...)## S3 method for class 'matchTab' print(x, ...)
x |
object of class 'matchTab' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'matchTab'
Print results for sample size for hypothesis testing of 2 means
## S3 method for class 'n.for.2means' print(x, ...)## S3 method for class 'n.for.2means' print(x, ...)
x |
object of class 'n.for.2means' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2p'
n.for.2means(mu1 = 10, mu2 = 14, sd1=3, sd2=3.5) n.for.2means(mu1 = 10, mu2 = 7:14, sd1=3, sd2=3.5) -> a an.for.2means(mu1 = 10, mu2 = 14, sd1=3, sd2=3.5) n.for.2means(mu1 = 10, mu2 = 7:14, sd1=3, sd2=3.5) -> a a
Print results for sample size for hypothesis testing of 2 proportions
## S3 method for class 'n.for.2p' print(x, ...)## S3 method for class 'n.for.2p' print(x, ...)
x |
object of class 'n.for.2p' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2p'
n.for.2p(p1=.1, p2=.2) n.for.2p(p1=seq(1,9,.5)/10, p2=.5)n.for.2p(p1=.1, p2=.2) n.for.2p(p1=seq(1,9,.5)/10, p2=.5)
Print results for sample size for hypothesis testing of 2 means in cluster RCT
## S3 method for class 'n.for.cluster.2means' print(x, ...)## S3 method for class 'n.for.cluster.2means' print(x, ...)
x |
object of class 'n.for.cluster.2means' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.cluster.2means'
Print results for sample size for hypothesis testing of 2 proportions in cluster RCT
## S3 method for class 'n.for.cluster.2p' print(x, ...)## S3 method for class 'n.for.cluster.2p' print(x, ...)
x |
object of class 'n.for.cluster.2p' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.cluster.2p'
Print results for sample size for hypothesis testing of 2 proportions in equivalent trial
## S3 method for class 'n.for.equi.2p' print(x, ...)## S3 method for class 'n.for.equi.2p' print(x, ...)
x |
object of class 'n.for.equi.2p' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2p'
n.for.equi.2p(p=.85, sig.diff=.05)n.for.equi.2p(p=.85, sig.diff=.05)
Print results for sample size for lot quality assurance sampling
## S3 method for class 'n.for.lqas' print(x, ...)## S3 method for class 'n.for.lqas' print(x, ...)
x |
object of class 'n.for.lqas' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
n.for.lqas(p0 = 0.05, q=0) n.for.lqas(p0 = (10:1)/100, q=0 ) -> a an.for.lqas(p0 = 0.05, q=0) n.for.lqas(p0 = (10:1)/100, q=0 ) -> a a
Print results for sample size for hypothesis testing of 2 proportions in non-inferior trial
## S3 method for class 'n.for.noninferior.2p' print(x, ...)## S3 method for class 'n.for.noninferior.2p' print(x, ...)
x |
object of class 'n.for.noninferior.2p' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2p'
n.for.noninferior.2p(p=.85, sig.inferior=.05)n.for.noninferior.2p(p=.85, sig.inferior=.05)
Print results for sample size of a continuous variable
## S3 method for class 'n.for.survey' print(x, ...)## S3 method for class 'n.for.survey' print(x, ...)
x |
object of class 'n.for.survey' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2p'
n.for.survey(p=seq(5,95,5)/100)n.for.survey(p=seq(5,95,5)/100)
Print results for power for hypothesis testing of 2 means
## S3 method for class 'power.for.2means' print(x, ...)## S3 method for class 'power.for.2means' print(x, ...)
x |
object of class 'power.for.2means' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2means'
power.for.2means(mu1 = 10, mu2=14, n1=5, n2=7, sd1=3, sd2=3.5) power.for.2means(mu1 = 10, mu2=7:14, n1=20, n2=25, sd1=3, sd2=3.5) -> a apower.for.2means(mu1 = 10, mu2=14, n1=5, n2=7, sd1=3, sd2=3.5) power.for.2means(mu1 = 10, mu2=7:14, n1=20, n2=25, sd1=3, sd2=3.5) -> a a
Print results for power of hypothesis testing of 2 proportions
## S3 method for class 'power.for.2p' print(x, ...)## S3 method for class 'power.for.2p' print(x, ...)
x |
object of class 'power.for.2p' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'n.for.2p'
power.for.2p(p1=.1, p2=.2, n1=10, n2=15) power.for.2p(p1=seq(1,9,.5)/10, p2=.5, n1=100, n2=120)power.for.2p(p1=.1, p2=.2, n1=10, n2=15) power.for.2p(p1=seq(1,9,.5)/10, p2=.5, n1=100, n2=120)
Print a statStack object
## S3 method for class 'statStack' print(x, ...)## S3 method for class 'statStack' print(x, ...)
x |
object of class 'statStack' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'statStack'
Print summary of data frame
## S3 method for class 'summ.data.frame' print(x, ...)## S3 method for class 'summ.data.frame' print(x, ...)
x |
object of class 'summ.data.frame' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'summ'
Print summary of a variable
## S3 method for class 'summ.default' print(x, ...)## S3 method for class 'summ.default' print(x, ...)
x |
object of class 'summ.default' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'summ'
Print a tableStack object
## S3 method for class 'tableStack' print(x, ...)## S3 method for class 'tableStack' print(x, ...)
x |
object of class 'tableStack' |
... |
further arguments passed to or used by methods. |
Virasakdi Chongsuvivatwong [email protected]
'tableStack'
data(Attitudes) tableStack(qa1:qa18, dataFrame=Attitudes) -> a print(a) data(Ectopic) tableStack(hia, gravi, by=outc, dataFrame=Ectopic) -> b print(b)data(Attitudes) tableStack(qa1:qa18, dataFrame=Attitudes) -> a print(a) data(Ectopic) tableStack(hia, gravi, by=outc, dataFrame=Ectopic) -> b print(b)
Create a population pyramid from age and sex
pyramid (age, sex, binwidth = 5, inputTable = NULL, printTable = FALSE, percent = "none", col.gender = NULL, bar.label = "auto", decimal = 1, col = NULL, cex.bar.value = 0.8, cex.axis = 1, main = "auto", cex.main = 1.2, ...)pyramid (age, sex, binwidth = 5, inputTable = NULL, printTable = FALSE, percent = "none", col.gender = NULL, bar.label = "auto", decimal = 1, col = NULL, cex.bar.value = 0.8, cex.axis = 1, main = "auto", cex.main = 1.2, ...)
age |
a numeric variable for age |
sex |
a variable of two levels for sexes, can be numeric but preferrably factor with labelled levels or characters |
binwidth |
bin width of age for each bar |
inputTable |
a table to read in with two columns of sexes and rows of age groups |
printTable |
whether the output table should be displayed on the console |
percent |
whether the lengths of the bars should be calculated from freqencies ("none" as default), percent of "each" sex or "total"" percentages |
col.gender |
vector reflecting colours of the two gender |
bar.label |
whether the bars would be labelled with the values |
decimal |
number of decimals displayed in the percent output table |
col |
colour(s) of the bars |
cex.bar.value |
character extension factor of the bar labels |
cex.axis |
character extension factor of the axis |
main |
main title |
cex.main |
character extension factor of main title |
... |
graph options for the bars, e.g. col |
'pyramid' draws a horizontal bar graph of age by sex.
The parameters of graph (par) options can be applied to 'font.lab' and those of the bars, e.g. 'col' but not of others.
Other lower level graph commands should be only for adding a 'title'.
'bar.label' when set as "auto", will be TRUE when 'percent="each"' or 'percent="total"'
When the variables age and sex are input arguments, the return object includes age group variable and the output table. The argument 'decimal' controls only decimals of the output displayed on the console but not the returned table.
Virasakdi Chongsuvivatwong [email protected]
'barplot', 'levels', 'table'
data(Oswego) .data <- Oswego attach(.data) pyramid(age, sex) pyramid(age, sex, bar.label = TRUE) pyramid(age, sex, printTable=TRUE) pyramid(age, sex, percent = "each", printTable=TRUE) pyramid(age, sex, percent = "total", printTable=TRUE) pyramid(age, sex, percent = "total", bar.label = FALSE) pyramid(age, sex, percent = "total", cex.bar.value = .5) pyramid(age, sex, col="red") pyramid(age, sex, col=1:16) # Too colorful! pyramid(age, sex, col.gender = c("pink","lightblue")) output <- pyramid(age, sex, binwidth = 10, percent="each", decimal=2) agegr <- output$ageGroup detach(.data) rm(list=ls()) # Drawing population pyramid from an exisiting table pyramid(inputTable=VADeaths[,1:2], font.lab=4) pyramid(inputTable=VADeaths[,1:2], font.lab=4, main=NULL) title("Death rates per 100 in rural Virginia in 1940")data(Oswego) .data <- Oswego attach(.data) pyramid(age, sex) pyramid(age, sex, bar.label = TRUE) pyramid(age, sex, printTable=TRUE) pyramid(age, sex, percent = "each", printTable=TRUE) pyramid(age, sex, percent = "total", printTable=TRUE) pyramid(age, sex, percent = "total", bar.label = FALSE) pyramid(age, sex, percent = "total", cex.bar.value = .5) pyramid(age, sex, col="red") pyramid(age, sex, col=1:16) # Too colorful! pyramid(age, sex, col.gender = c("pink","lightblue")) output <- pyramid(age, sex, binwidth = 10, percent="each", decimal=2) agegr <- output$ageGroup detach(.data) rm(list=ls()) # Drawing population pyramid from an exisiting table pyramid(inputTable=VADeaths[,1:2], font.lab=4) pyramid(inputTable=VADeaths[,1:2], font.lab=4, main=NULL) title("Death rates per 100 in rural Virginia in 1940")
Display of various epidemiological modelling results in a medically understandable format
logistic.display(logistic.model, alpha = 0.05, crude = TRUE, crude.p.value = FALSE, decimal = 2, simplified = FALSE) clogistic.display(clogit.model, alpha = 0.05, crude=TRUE, crude.p.value=FALSE, decimal = 2, simplified = FALSE) cox.display (cox.model, alpha = 0.05, crude=TRUE, crude.p.value=FALSE, decimal = 2, simplified = FALSE) regress.display(regress.model, alpha = 0.05, crude = FALSE, crude.p.value = FALSE, decimal = 2, simplified = FALSE) idr.display(idr.model, alpha = 0.05, crude = TRUE, crude.p.value = FALSE, decimal = 2, simplified = FALSE) mlogit.display(multinom.model, decimal = 2, alpha = 0.05) ordinal.or.display(ordinal.model, decimal = 3, alpha = 0.05) tableGlm (model, modified.coeff.array, decimal) ## S3 method for class 'display' print(x, ...)logistic.display(logistic.model, alpha = 0.05, crude = TRUE, crude.p.value = FALSE, decimal = 2, simplified = FALSE) clogistic.display(clogit.model, alpha = 0.05, crude=TRUE, crude.p.value=FALSE, decimal = 2, simplified = FALSE) cox.display (cox.model, alpha = 0.05, crude=TRUE, crude.p.value=FALSE, decimal = 2, simplified = FALSE) regress.display(regress.model, alpha = 0.05, crude = FALSE, crude.p.value = FALSE, decimal = 2, simplified = FALSE) idr.display(idr.model, alpha = 0.05, crude = TRUE, crude.p.value = FALSE, decimal = 2, simplified = FALSE) mlogit.display(multinom.model, decimal = 2, alpha = 0.05) ordinal.or.display(ordinal.model, decimal = 3, alpha = 0.05) tableGlm (model, modified.coeff.array, decimal) ## S3 method for class 'display' print(x, ...)
logistic.model |
a model from a logistic regression |
clogit.model |
a model from a conditional logistic regression |
regress.model |
a model from a linear regression |
cox.model |
a model from a cox regression |
idr.model |
a model from a Poisson regression or a negative binomial regression |
multinom.model |
a model from a multinomial or polytomous regression |
ordinal.model |
a model from an ordinal logistic regression |
alpha |
significance level |
crude |
whether crude results and their confidence intervals should also be displayed |
crude.p.value |
whether crude P values should also be displayed if and only if 'crude=TRUE' |
decimal |
number of decimal places displayed |
simplified |
whether the display should be simplified |
model |
model passed from logistic.display or regress.display to tableGlm |
modified.coeff.array |
array of model coefficients sent to the function 'tableGlm' to produce the final output |
x |
object obtained from these 'display' functions |
... |
further arguments passed to or used by methods |
R provides several epidemiological modelling techniques. The functions above display these results in a format easier for medical people to understand.
The function 'tableGlm' is not for general use. It is called by other display functions to receive the 'modified.coeff.array' and produce the output table.
The argument 'simplified' has a default value of 'FALSE'. It works best if the 'data' argument has been supplied during creation of the model. Under this condition, the output has three parts. Part 1 (the first line) indicates the type of the regression and the outcome. For logistic regression, if the outcome is a factor then the referent level is shown. Part 2 shows the main output table where each independent variable coefficient is displayed. If the independent variable is continuous (class numeric) then name of the variable is shown (or the descriptive label if it exists). If the variable is a factor then the name of the level is shown with the referent level omitted. In this case, the name of the referent level and the statistic testing for group effects are displayed. An F-test is used when the model is of class 'lm' or 'glm' with 'family=gaussian' specified. A Likelihood Ratio test is performed when the model is of class 'glm' with 'family = binomial' or 'family = poisson' specified and for models of class 'coxph' and 'clogit'. These tests are carried out with the records available in the model, not necessary all records in the full 'data' argument. The number of records in the model is displayed in the part 3 of the output. When 'simplified=TRUE', the first and the last parts are omitted from the display.
The result is an object of class 'display' and 'list'. Their apparence on the R console is controlled by 'print.display'. The 'table' attribute of these 'display' objects are ready to write (using 'write.csv') to a .csv file which can then be copied to a manuscript document. This approach can substantially reduce both the time and errors produced due to conventional manual copying.
'logistic.display', 'regress.display', 'clogit.display' and 'cox.display', each produces an output table. See 'details'.
Before using these 'display' functions, please note the following limitations.
1) Users should define the 'data' argument of the model.
2) The names of the independent variables must be a subset of the names of the variables in the 'data' argument. Sometimes, one of more variables are omitted by the model due to collinearity. In such a case, users have to specify 'simplified=TRUE' in order to get the display function to work.
3) Under the following conditions, 'simplified' will be forced to TRUE and 'crude' forced to FALSE.
3.1) The names of the independent variables contain a function such as 'factor()' or any '\$' sign.
3.2) The levels of the factor variables contain any ':' sign.
3.3) There are more than one interaction terms in the model
3.4) The 'data' argument is missing in the conditional logistic regression and Cox regression model
4) For any other problems with these display results, users are advised to run 'summary(model)' or 'summary(model)$coefficients' to check the consistency between variable names in the model and those in the coefficients. The number in the latter may be fewer than that in the former due to collinearity. In this case, it is advised to specify 'simplified=TRUE' to turn off the attempt to tidy up the rownames of the output from 'summary(model)$coeffients'. The output when 'simplified=TRUE' is more reliable but less understandable.
Virasakdi Chongsuvivatwong [email protected]
'glm', 'confint'
model0 <- glm(case ~ induced + spontaneous, family=binomial, data=infert) summary(model0) logistic.display(model0) data(ANCdata) glm1 <- glm(death ~ anc + clinic, family=binomial, data=ANCdata) logistic.display(glm1) logistic.display(glm1, simplified=TRUE) library(MASS) # necessary for negative binomial regression data(DHF99); .data <- DHF99 attach(.data) model.poisson <- glm(containers ~ education + viltype, family=poisson, data=DHF99) model.nb <- glm.nb(containers ~ education + viltype, data=.data) idr.display(model.poisson) -> poiss print(poiss) # or print.display(poiss) or poiss idr.display(model.nb) detach(.data) data(VC1to6) .data <- VC1to6 .data$fsmoke <- factor(.data$smoking) levels(.data$fsmoke) <- list("no"=0, "yes"=1) clr1 <- clogit(case ~ alcohol + fsmoke + strata(matset), data=.data) clogistic.display(clr1) rm(list=ls()) data(BP) .data <- BP attach(.data) age <- as.numeric(as.Date("2000-01-01") - birthdate)/365.25 agegr <- pyramid(age,sex, bin=20)$ageGroup .data$hypertension <- sbp >= 140 | dbp >=90 detach(.data) model1 <- glm(hypertension ~ sex + agegr + saltadd, family=binomial, data=.data) logistic.display(model1) -> table3 attributes(table3) table3 table3$table # You may want to save table3 into a spreadsheet write.csv(table3$table, file="table3.csv") # Note $table ## Have a look at this file in Excel, or similar spreadsheet program file.remove(file="table3.csv") model2 <- glm(hypertension ~ sex * age + sex * saltadd, family=binomial, data=.data) logistic.display(model2) # More than 1 interaction term so 'simplified turned to TRUE reg1 <- lm(sbp ~ sex + agegr + saltadd, data=.data) regress.display(reg1) reg2 <- glm(sbp ~ sex + agegr + saltadd, family=gaussian, data=.data) regress.display(reg2) data(Compaq) cox1 <- coxph(Surv(year, status) ~ hospital + stage * ses, data=Compaq) cox.display(cox1, crude.p.value=TRUE) # Ordinal logistic regression library(nnet) options(contrasts = c("contr.treatment", "contr.poly")) house.plr <- polr(Sat ~ Infl + Type + Cont, weights = Freq, data = housing) house.plr ordinal.or.display(house.plr) # Polytomous or multinomial logistic regression house.multinom <- multinom(Sat ~ Infl + Type + Cont, weights = Freq, data = housing) summary(house.multinom) mlogit.display(house.multinom, alpha=.01) # with 99% confidence limits.model0 <- glm(case ~ induced + spontaneous, family=binomial, data=infert) summary(model0) logistic.display(model0) data(ANCdata) glm1 <- glm(death ~ anc + clinic, family=binomial, data=ANCdata) logistic.display(glm1) logistic.display(glm1, simplified=TRUE) library(MASS) # necessary for negative binomial regression data(DHF99); .data <- DHF99 attach(.data) model.poisson <- glm(containers ~ education + viltype, family=poisson, data=DHF99) model.nb <- glm.nb(containers ~ education + viltype, data=.data) idr.display(model.poisson) -> poiss print(poiss) # or print.display(poiss) or poiss idr.display(model.nb) detach(.data) data(VC1to6) .data <- VC1to6 .data$fsmoke <- factor(.data$smoking) levels(.data$fsmoke) <- list("no"=0, "yes"=1) clr1 <- clogit(case ~ alcohol + fsmoke + strata(matset), data=.data) clogistic.display(clr1) rm(list=ls()) data(BP) .data <- BP attach(.data) age <- as.numeric(as.Date("2000-01-01") - birthdate)/365.25 agegr <- pyramid(age,sex, bin=20)$ageGroup .data$hypertension <- sbp >= 140 | dbp >=90 detach(.data) model1 <- glm(hypertension ~ sex + agegr + saltadd, family=binomial, data=.data) logistic.display(model1) -> table3 attributes(table3) table3 table3$table # You may want to save table3 into a spreadsheet write.csv(table3$table, file="table3.csv") # Note $table ## Have a look at this file in Excel, or similar spreadsheet program file.remove(file="table3.csv") model2 <- glm(hypertension ~ sex * age + sex * saltadd, family=binomial, data=.data) logistic.display(model2) # More than 1 interaction term so 'simplified turned to TRUE reg1 <- lm(sbp ~ sex + agegr + saltadd, data=.data) regress.display(reg1) reg2 <- glm(sbp ~ sex + agegr + saltadd, family=gaussian, data=.data) regress.display(reg2) data(Compaq) cox1 <- coxph(Surv(year, status) ~ hospital + stage * ses, data=Compaq) cox.display(cox1, crude.p.value=TRUE) # Ordinal logistic regression library(nnet) options(contrasts = c("contr.treatment", "contr.poly")) house.plr <- polr(Sat ~ Infl + Type + Cont, weights = Freq, data = housing) house.plr ordinal.or.display(house.plr) # Polytomous or multinomial logistic regression house.multinom <- multinom(Sat ~ Infl + Type + Cont, weights = Freq, data = housing) summary(house.multinom) mlogit.display(house.multinom, alpha=.01) # with 99% confidence limits.
Receiver Operating Characteristic curve of a logistic regression model and a diagnostic table
lroc(logistic.model, graph = TRUE, add = FALSE, title = FALSE, line.col = "red", auc.coords = NULL, grid = TRUE, grid.col = "blue", ...) roc.from.table(table, graph = TRUE, add = FALSE, title = FALSE, line.col = "red", auc.coords = NULL, grid = TRUE, grid.col = "blue", ...)lroc(logistic.model, graph = TRUE, add = FALSE, title = FALSE, line.col = "red", auc.coords = NULL, grid = TRUE, grid.col = "blue", ...) roc.from.table(table, graph = TRUE, add = FALSE, title = FALSE, line.col = "red", auc.coords = NULL, grid = TRUE, grid.col = "blue", ...)
logistic.model |
A model from logistic regression |
table |
A cross tabulation of the levels of a test (rows) vs a gold standard positive and negative (columns) |
graph |
Draw ROC curve |
add |
Whether the line is drawn on the existing ROC curve |
title |
If true, the model will be displayed as main title |
line.col |
Color of the line |
auc.coords |
Coordinates for label of 'auc' (area under curve) |
grid |
Whether the grid should be drawn |
grid.col |
Grid colour, if drawn |
... |
Additional graphic parameters |
'lroc' graphs the ROC curve of a logistic regression model. If ‘table=TRUE’, the diagnostic table based on the regression will be printed out.
'roc.from.table' computes the change of sensitivity and specificity of each cut point and uses these for drawing the ROC curve.
In both cases, the area under the curve is computed.
Virasakdi Chongsuvivatwong [email protected]
'glm'
# Single ROC curve from logistic regression # Note that 'induced' and 'spontaneous' are both originally continuous variables model1 <- glm(case ~ induced + spontaneous, data=infert, family=binomial) logistic.display(model1) # Having two spontaneous abortions is quite close to being infertile! # This is actually not a causal relationship lroc(model1, title=TRUE, auc.coords=c(.5,.1)) # For PowerPoint presentation, the graphic elements should be enhanced as followed lroc(model1, title=TRUE, cex.main=2, cex.lab=1.5, col.lab="blue", cex.axis=1.3, lwd=3) lroc1 <- lroc(model1) # The main title and auc text have disappeared model2 <- glm(case ~ spontaneous, data=infert, family=binomial) logistic.display(model2) lroc2 <- lroc(model2, add=TRUE, line.col="brown", lty=2) legend("bottomright",legend=c(lroc1$model.description, lroc2$model.description), lty=1:2, col=c("red","brown"), bg="white") title(main="Comparison of two logistic regression models") lrtest(model1, model2) # Number of induced abortions is associated with increased risk for infertility # Various form of logistic regression # Case by case data data(ANCdata) .data <- ANCdata glm1 <- glm(death ~ anc + clinic, binomial, data=.data) # Note 'calc' lroc(glm1) # Frequency format data(ANCtable) ANCtable .data <- ANCtable attach(.data) death <- factor (death) levels (death) <- c("no","yes") anc <- with(.data, factor (anc)) levels (anc) <- c("old","new") clinic <- with(.data, factor (clinic)) levels (clinic) <- c("A","B") .data <- data.frame(death, anc, clinic) .data glm2 <- glm(death ~ anc + clinic, binomial, weights=Freq, data=.data) lroc(glm2) detach(.data) # ROC from a diagnostic table table1 <- as.table(cbind(c(1,27,56,15,1),c(0,0,10,69,21))) colnames(table1) <- c("Non-diseased", "Diseased") rownames(table1) <- c("15-29","30-44","45-59","60-89","90+") table1 roc.from.table(table1) roc.from.table(table1, title=TRUE, auc.coords=c(.4,.1), cex=1.2) # Application of the returned list roc1 <- roc.from.table(table1, graph=FALSE) cut.points <- rownames(roc1$diagnostic.table) text(x=roc1$diagnostic.table[,1], y=roc1$diagnostic.table[,2], labels=cut.points, cex=1.2, col="brown") rm(list=ls())# Single ROC curve from logistic regression # Note that 'induced' and 'spontaneous' are both originally continuous variables model1 <- glm(case ~ induced + spontaneous, data=infert, family=binomial) logistic.display(model1) # Having two spontaneous abortions is quite close to being infertile! # This is actually not a causal relationship lroc(model1, title=TRUE, auc.coords=c(.5,.1)) # For PowerPoint presentation, the graphic elements should be enhanced as followed lroc(model1, title=TRUE, cex.main=2, cex.lab=1.5, col.lab="blue", cex.axis=1.3, lwd=3) lroc1 <- lroc(model1) # The main title and auc text have disappeared model2 <- glm(case ~ spontaneous, data=infert, family=binomial) logistic.display(model2) lroc2 <- lroc(model2, add=TRUE, line.col="brown", lty=2) legend("bottomright",legend=c(lroc1$model.description, lroc2$model.description), lty=1:2, col=c("red","brown"), bg="white") title(main="Comparison of two logistic regression models") lrtest(model1, model2) # Number of induced abortions is associated with increased risk for infertility # Various form of logistic regression # Case by case data data(ANCdata) .data <- ANCdata glm1 <- glm(death ~ anc + clinic, binomial, data=.data) # Note 'calc' lroc(glm1) # Frequency format data(ANCtable) ANCtable .data <- ANCtable attach(.data) death <- factor (death) levels (death) <- c("no","yes") anc <- with(.data, factor (anc)) levels (anc) <- c("old","new") clinic <- with(.data, factor (clinic)) levels (clinic) <- c("A","B") .data <- data.frame(death, anc, clinic) .data glm2 <- glm(death ~ anc + clinic, binomial, weights=Freq, data=.data) lroc(glm2) detach(.data) # ROC from a diagnostic table table1 <- as.table(cbind(c(1,27,56,15,1),c(0,0,10,69,21))) colnames(table1) <- c("Non-diseased", "Diseased") rownames(table1) <- c("15-29","30-44","45-59","60-89","90+") table1 roc.from.table(table1) roc.from.table(table1, title=TRUE, auc.coords=c(.4,.1), cex=1.2) # Application of the returned list roc1 <- roc.from.table(table1, graph=FALSE) cut.points <- rownames(roc1$diagnostic.table) text(x=roc1$diagnostic.table[,1], y=roc1$diagnostic.table[,2], labels=cut.points, cex=1.2, col="brown") rm(list=ls())
Sample size calculations for epidemiological studies
n.for.survey (p, delta = "auto", popsize = NULL, deff = 1, alpha = 0.05) n.for.2means (mu1, mu2, sd1, sd2, ratio = 1, alpha = 0.05, power = 0.8) n.for.cluster.2means (mu1, mu2, sd1, sd2, alpha = 0.05, power = 0.8, ratio = 1, mean.cluster.size = 10, previous.mean.cluster.size = NULL, previous.sd.cluster.size = NULL, max.cluster.size = NULL, min.cluster.size = NULL, icc = 0.1) n.for.2p (p1, p2, alpha = 0.05, power = 0.8, ratio = 1) n.for.cluster.2p (p1, p2, alpha = 0.05, power = 0.8, ratio = 1, mean.cluster.size = 10, previous.mean.cluster.size = NULL, previous.sd.cluster.size = NULL, max.cluster.size = NULL, min.cluster.size = NULL, icc = 0.1) n.for.equi.2p(p, sig.diff, alpha=.05, power=.8) n.for.noninferior.2p (p, sig.inferior, alpha = 0.05, power = 0.8) n.for.lqas (p0, q = 0, N = 10000, alpha = 0.05, exact = FALSE)n.for.survey (p, delta = "auto", popsize = NULL, deff = 1, alpha = 0.05) n.for.2means (mu1, mu2, sd1, sd2, ratio = 1, alpha = 0.05, power = 0.8) n.for.cluster.2means (mu1, mu2, sd1, sd2, alpha = 0.05, power = 0.8, ratio = 1, mean.cluster.size = 10, previous.mean.cluster.size = NULL, previous.sd.cluster.size = NULL, max.cluster.size = NULL, min.cluster.size = NULL, icc = 0.1) n.for.2p (p1, p2, alpha = 0.05, power = 0.8, ratio = 1) n.for.cluster.2p (p1, p2, alpha = 0.05, power = 0.8, ratio = 1, mean.cluster.size = 10, previous.mean.cluster.size = NULL, previous.sd.cluster.size = NULL, max.cluster.size = NULL, min.cluster.size = NULL, icc = 0.1) n.for.equi.2p(p, sig.diff, alpha=.05, power=.8) n.for.noninferior.2p (p, sig.inferior, alpha = 0.05, power = 0.8) n.for.lqas (p0, q = 0, N = 10000, alpha = 0.05, exact = FALSE)
p |
estimated probability |
delta |
difference between the estimated prevalence and one side of the 95 percent confidence limit (precision) |
popsize |
size of the finite population |
deff |
design effect for cluster sampling |
alpha |
significance level |
mu1, mu2
|
estimated means of the two populations |
sd1, sd2
|
estimated standard deviations of the two populations |
ratio |
n2/n1 |
mean.cluster.size |
mean of the cluster size planned in the current study |
previous.mean.cluster.size, previous.sd.cluster.size
|
mean and sd of cluster size from a previous study |
max.cluster.size, min.cluster.size
|
maximum and minimum of cluster size in the current study |
icc |
intraclass correlation coefficient |
p1, p2
|
estimated probabilities of the two populations |
power |
power of the study |
sig.diff |
level of difference consider as being clinically significant |
sig.inferior |
level of reduction of effectiveness as being clinically significant |
p0 |
critical proportion beyond which the lot will be rejected |
q |
critical number of faulty pieces found in the sample, beyond which the lot will be rejected |
N |
lot size |
exact |
whether the exact probability is to be computed |
'n.for.survey' is used to compute the sample size required to conduct a survey.
When 'delta="auto"', delta will change according to the value of p. If 0.3 <= p <= 0.7, delta = 0.1. If 0.1 <= p < .3, or 0.7< p <=0.9, then delta=.05. Finally, if p < 0.1, then delta = p/2. If 0.9 < p, then delta = (1-p)/2.
When cluster sampling is employed, the design effect (deff) has to be taken into account.
'n.for.2means' is used to compute the sample size needed for testing the hypothesis that the difference of two population means is zero.
'n.for.cluster.2means' and 'n.for.cluster.2p' are for cluster (usually randomized) controlled trial.
'n.for.2p' is used to the compute the sample size needed for testing the hypothesis that the difference of two population proportions is zero.
'n.for.equi.2p' is used for equivalent trial with equal probability of success or fail being p for both groups. 'sig.diff' is a difference in probability considered as being clinically significant. If both sides of limits of 95 percent CI of the difference are within +sig.diff or -sig.diff, there would be neither evidence of inferiority nor of superiority of any arm.
'n.for.noninferior.2p' is similar to 'n.for.equi.2p' except if the lower limit of 95 percent CI of the difference is higher than the sig.inferior level, the hypothesis of inferiority would be rejected.
For a case control study, p1 and p2 are the proportions of exposure among cases and controls.
For a cohort study, p1 and p2 are proportions of positive outcome among the exposed and non-exposed groups.
'ratio' in a case control study is controls:case. In cohort and cross-sectional studies, it is non-exposed:exposed.
LQAS stands for Lot Quality Assurance Sampling. The sample size n is determined to test whether the lot of a product has a defective proportion exceeding a critical proportion, p0. Out of the sample tested, if the number of defective specimens is greater than q, the lot is considered not acceptable. This concept can be applied to quality assurance processes in health care.
When any parameter is a vector of length > 5, a table of sample size by the varying values of parameters is displayed.
a list.
'n.for.survey' returns an object of class "n.for.survey"
'n.for.2p' returns an object of class "n.for.2p"
'n.for.2means' returns an object of class "n.for.2means"
'n.for.lqas' returns an object of class "n.for.lqas"
Each type of returned values consists of vectors of various parameters in the formula and the required sample size(s).
Virasakdi Chongsuvivatwong [email protected]
Eldridge SM, Ashby D, Kerry S. 2006 Sample size for cluster randomized trials: effect of coefficient of variation of cluster size and analysis method. Int J Epidemiol 35(5): 1292-300.
'power.for.2means', 'power.for.2p'
# In a standard survey to determine the coverage of immunization needed using # a cluster sampling technique on a population of approximately 500000, and # an estimated prevalence of 70 percent, design effect is assumed to be 2. n.for.survey( p = .8, delta = .1, popsize = 500000, deff =2) # 123 needed # To see the effect of prevalence on delta and sample size n.for.survey( p = c(.5, .6, .7, .8, .9, .95, .99)) # Testing the efficacy of measles vaccine in a case control study . # The coverage in the non-diseased population is estimated at 80 percent. # That in the diseased is 60 percent. n.for.2p(p1=.8, p2=.6) # n1=n2=91 needed # A randomized controlled trial testing cure rate of a disease of # 90 percent by new drugs and 80 percent by the old one. n.for.2p(p1=.9, p2=.8) # 219 subjects needed in each arm. # To see the effect of p1 on sample size n.for.2p(p1=seq(1,9,.5)/10, p2=.5) # A table output # The same randomized trial to check whether the new treatment is 5 percent # different from the standard treatment assuming both arms has a common # cure rate of 85 percent would be n.for.equi.2p(p=.85, sig.diff=0.05) # 801 each. # If inferior arm is not allow to be lower than -0.05 (5 percent less effective) n.for.noninferior.2p(p=.85, sig.inferior=0.05) # 631 each. # A cluster randomized controlled trial to test whether training of village # volunteers would result in reduction of prevalence of a disease from 50 percent # in control villages to 30 percent in the study village with a cluster size # varies from 250 to 500 eligible subjects per village (mean of 350) and the # intraclass correlation is assumed to be 0.15 n.for.cluster.2p(p1=.5, p2=.3, mean.cluster.size = 350, max.cluster.size = 500, min.cluster.size = 250, icc = 0.15) # A quality assurance to check whether the coding of ICD-10 is faulty # by no more than 2 percent.The minimum sample is required. # Thus any faulty coding in the sample is not acceptable. n.for.lqas(p0 = .02, q=0, exact = TRUE) # 148 non-faulty checks is required # to support the assurance process. n.for.lqas(p0 = (1:10)/100, q=0, exact = FALSE)# In a standard survey to determine the coverage of immunization needed using # a cluster sampling technique on a population of approximately 500000, and # an estimated prevalence of 70 percent, design effect is assumed to be 2. n.for.survey( p = .8, delta = .1, popsize = 500000, deff =2) # 123 needed # To see the effect of prevalence on delta and sample size n.for.survey( p = c(.5, .6, .7, .8, .9, .95, .99)) # Testing the efficacy of measles vaccine in a case control study . # The coverage in the non-diseased population is estimated at 80 percent. # That in the diseased is 60 percent. n.for.2p(p1=.8, p2=.6) # n1=n2=91 needed # A randomized controlled trial testing cure rate of a disease of # 90 percent by new drugs and 80 percent by the old one. n.for.2p(p1=.9, p2=.8) # 219 subjects needed in each arm. # To see the effect of p1 on sample size n.for.2p(p1=seq(1,9,.5)/10, p2=.5) # A table output # The same randomized trial to check whether the new treatment is 5 percent # different from the standard treatment assuming both arms has a common # cure rate of 85 percent would be n.for.equi.2p(p=.85, sig.diff=0.05) # 801 each. # If inferior arm is not allow to be lower than -0.05 (5 percent less effective) n.for.noninferior.2p(p=.85, sig.inferior=0.05) # 631 each. # A cluster randomized controlled trial to test whether training of village # volunteers would result in reduction of prevalence of a disease from 50 percent # in control villages to 30 percent in the study village with a cluster size # varies from 250 to 500 eligible subjects per village (mean of 350) and the # intraclass correlation is assumed to be 0.15 n.for.cluster.2p(p1=.5, p2=.3, mean.cluster.size = 350, max.cluster.size = 500, min.cluster.size = 250, icc = 0.15) # A quality assurance to check whether the coding of ICD-10 is faulty # by no more than 2 percent.The minimum sample is required. # Thus any faulty coding in the sample is not acceptable. n.for.lqas(p0 = .02, q=0, exact = TRUE) # 148 non-faulty checks is required # to support the assurance process. n.for.lqas(p0 = (1:10)/100, q=0, exact = FALSE)
Setting locale and internationalizing Epicalc graph title
setTitle(locale)setTitle(locale)
locale |
A string denoting international language of choice |
On calling 'library(epicalc)', '.locale()' has an inital value of FALSE, ie. the titles of Epicalc's automatic graphs are displayed in the English language. 'setTitle' has two effects. It selects the locale and resets the hidden object '.locale()' to TRUE. The command internationalizes the title of automatic graphs created by Epicalc according to 'locale' given in the function's argument.
If '.locale()' is TRUE, then the automatic graphs produced by Epicalc commands, such as 'summ(var)' or 'tab1(var)' or 'tabpct(var1,var2)', will lookup a language conversion table for the graph title and the title will be changed accordingly.
Internationalization of the title can be disabled by typing '.locale(FALSE)'. This has no effect of locale as a whole unless it is reset to English by issuing the command 'setTitle("English")'.
Virasakdi Chongsuvivatwong [email protected]
'Sys.setlocale', 'Sys.getlocale' and 'titleString'
.data <- iris attach(.data) summ(Sepal.Length, by=Species) setTitle("English") dotplot(Sepal.Length, by=Species) setTitle("Malay") dotplot(Sepal.Length, by=Species) setTitle("Spanish") dotplot(Sepal.Length, by=Species) detach(.data) rm(.data).data <- iris attach(.data) summ(Sepal.Length, by=Species) setTitle("English") dotplot(Sepal.Length, by=Species) setTitle("Malay") dotplot(Sepal.Length, by=Species) setTitle("Spanish") dotplot(Sepal.Length, by=Species) detach(.data) rm(.data)
Quantile-normal plots with Shapiro-Wilk's test result integrated
shapiro.qqnorm (x, ...)shapiro.qqnorm (x, ...)
x |
A numeric vector |
... |
Graphical parameters passed to 'par' |
To test a variable 'x' against the normal distribution, a qqnorm plot is integrated with the Shapiro-Wilk test to enhance interpretation.
Virasakdi Chongsuvivatwong [email protected]
'shapiro.test', 'qqnorm', 'par'
x <- rnorm(10) a <- LETTERS[1:10] shapiro.qqnorm(x, pch=a, col="red") qqline(x, lty=2, col="black")x <- rnorm(10) a <- LETTERS[1:10] shapiro.qqnorm(x, pch=a, col="red") qqline(x, lty=2, col="black")
Sleepiness among participants in a workshop
data(Sleep3)data(Sleep3)
A data frame with 15 observations on the following 8 variables.
ida numeric vector
gendera factor with levels male female
dbirtha Date vector for birth date
sleepya numeric vector for any experience of sleepiness in the class: 0=no 1=yes
lecturea numeric vector for ever felt sleepy during a lecture: 0=no 1=yes
grworka numeric vector for ever felt sleepy during a group work: 0=no 1=yes
kga numeric vector
cma numeric vector
data(Sleep3) des(Sleep3)data(Sleep3) des(Sleep3)
Compares the difference in means or medians of the levels of a factor or list of factors
statStack (cont.var, by, dataFrame, iqr="auto", var.labels = TRUE, decimal = 1, assumption.p.value = .01)statStack (cont.var, by, dataFrame, iqr="auto", var.labels = TRUE, decimal = 1, assumption.p.value = .01)
cont.var |
a single continuous variable in the data frame |
by |
a factor, or list of factors, each of length <code>nrow(dataFrame)</code> |
iqr |
to display median and inter-quartile range instead of mean and sd. |
var.labels |
use descriptions of the 'by' variables if available |
dataFrame |
source data frame of the variables |
decimal |
number of digits displayed after decimal point |
assumption.p.value |
level of Bartlett's test P value to judge whether the comparison and the test should be parametric |
This function computes means/medians of a continuous variable in each level of the specified factor(s) and performs an appropriate statistical test.
The classes of the variable to compute statistics must be either 'integer' or 'numeric' why all 'by' variables must be 'factor'.
Like in 'tableStack', the argument 'iqr' has a default value being "auto". Non-parametric comparison and test will be automatically chosen if Bartlette's test P value is below the 'assumption.p.value'.Like in 'tableStack', the default value for the 'iqr' argument is "auto", which means non-parametric comparison and test will be automatically chosen if the P-value from Bartlett's test is below the value of the 'assumption.p.value' argument (0.01).
The user can force the function to perform a parametric test by setting 'iqr=NULL' and to perform a non-parametric test by setting 'iqr' to the name or index of the continuous variable.
By default, 'var.labels=TRUE' in order to give nice output.
an object of class 'statStack' and 'table'
Virasakdi Chongsuvivatwong [email protected]
'tableStack'
statStack(Price, by=c(DriveTrain, Origin), dataFrame=Cars93) statStack(Price, by=c(DriveTrain, Origin), dataFrame=Cars93, iqr=NULL) Cars93$log10.Price <- log10(Cars93$Price)# added as the 28th variable statStack(log10.Price, by=c(DriveTrain, Origin), dataFrame=Cars93) statStack(log10.Price, by=c(DriveTrain, Origin), dataFrame=Cars93, iqr=28) rm(Cars93) data(Compaq) statStack(year, by=c(hospital, stage:ses), dataFrame=Compaq) # Note that var.labels 'Age group' is displayed instead of var. name 'agegr'statStack(Price, by=c(DriveTrain, Origin), dataFrame=Cars93) statStack(Price, by=c(DriveTrain, Origin), dataFrame=Cars93, iqr=NULL) Cars93$log10.Price <- log10(Cars93$Price)# added as the 28th variable statStack(log10.Price, by=c(DriveTrain, Origin), dataFrame=Cars93) statStack(log10.Price, by=c(DriveTrain, Origin), dataFrame=Cars93, iqr=28) rm(Cars93) data(Compaq) statStack(year, by=c(hospital, stage:ses), dataFrame=Compaq) # Note that var.labels 'Age group' is displayed instead of var. name 'agegr'
Summary of data frame in a convenient table. Summary of a variable with statistics and graph
summ(x, ...) ## Default S3 method: summ(x, by=NULL, graph = TRUE, box = FALSE, pch = 18, ylab = "auto", main = "auto", cex.X.axis = 1, cex.Y.axis = 1, dot.col = "auto", ...) ## S3 method for class 'factor' summ(x, by=NULL, graph=TRUE, ...) ## S3 method for class 'logical' summ(x, by=NULL, graph=TRUE, ...) ## S3 method for class 'data.frame' summ(x, ...)summ(x, ...) ## Default S3 method: summ(x, by=NULL, graph = TRUE, box = FALSE, pch = 18, ylab = "auto", main = "auto", cex.X.axis = 1, cex.Y.axis = 1, dot.col = "auto", ...) ## S3 method for class 'factor' summ(x, by=NULL, graph=TRUE, ...) ## S3 method for class 'logical' summ(x, by=NULL, graph=TRUE, ...) ## S3 method for class 'data.frame' summ(x, ...)
x |
'x' can be a data frame or a vector |
by |
a stratification variable, valid only when x is a vector |
graph |
automatic plot (sorted dot chart) if 'x' is a vector |
box |
add a boxplot to the graph (by=NULL) |
pch |
plot characters |
ylab |
annotation on Y axis |
main |
main title of the graph |
cex.X.axis |
character extension scale of X axis |
cex.Y.axis |
character extension scale of Y axis |
dot.col |
colour(s) of plot character(s) |
... |
additional graph parameters |
For data frames, 'summ' gives basic statistics of each variable in the data frame. The other arguments are ignored.
For single vectors, a sorted dot chart is also provided, if graph=TRUE (default).
Virasakdi Chongsuvivatwong [email protected]
'summary', 'use', 'des'
data(Oswego) .data <- Oswego summ(.data) with(.data, summ(age)) with(.data, summ(age, box=TRUE)) with(.data, summ(age, dot.col="brown")) with(.data, summ(age, by=sex)) # Changing dot colours with(.data, summ(age, by=sex, dot.col = c("blue","orange"))) # Enlarging main title and other elements with(.data, summ(age, by=sex, cex.main=1.5, cex.X.axis=1.5, cex.Y.axis=1.7)) # Free vector summ(rnorm(1000)) summ((1:100)^2, by=rep(1:2, 50)) summ((1:100)^2, by=rep(c("Odd","Even"), 50), main="Quadratic distribution by odd and even numbers")data(Oswego) .data <- Oswego summ(.data) with(.data, summ(age)) with(.data, summ(age, box=TRUE)) with(.data, summ(age, dot.col="brown")) with(.data, summ(age, by=sex)) # Changing dot colours with(.data, summ(age, by=sex, dot.col = c("blue","orange"))) # Enlarging main title and other elements with(.data, summ(age, by=sex, cex.main=1.5, cex.X.axis=1.5, cex.Y.axis=1.7)) # Free vector summ(rnorm(1000)) summ((1:100)^2, by=rep(1:2, 50)) summ((1:100)^2, by=rep(c("Odd","Even"), 50), main="Quadratic distribution by odd and even numbers")
One-way tabulation with automatic bar chart and optional indicator variables generation
tab1(x0, decimal = 1, sort.group = FALSE, cum.percent = !any(is.na(x0)), graph = TRUE, missing = TRUE, bar.values = "frequency", horiz = FALSE, cex = 1, cex.names = 1, main = "auto", xlab = "auto", ylab = "auto", col = "auto", gen.ind.vars = FALSE, ...) ## S3 method for class 'tab1' print(x, ...)tab1(x0, decimal = 1, sort.group = FALSE, cum.percent = !any(is.na(x0)), graph = TRUE, missing = TRUE, bar.values = "frequency", horiz = FALSE, cex = 1, cex.names = 1, main = "auto", xlab = "auto", ylab = "auto", col = "auto", gen.ind.vars = FALSE, ...) ## S3 method for class 'tab1' print(x, ...)
x0 |
a variable |
decimal |
number of decimals for the percentages in the table |
sort.group |
pattern for sorting categories in the table and in the chart. Default is no sorting. I can also be "decreasing" or "increasing". |
cum.percent |
presence of cumulative percentage in the output table. Default is TRUE for a variable without any missing values. |
graph |
whether a graph should be shown |
missing |
include the missing values category or <NA> in the graphic display |
bar.values |
include the value of frequency. This can also be "percent" or "none" at the end of each bar |
horiz |
set the bar chart to horizontal orientation |
cex |
parameter for extension of characters or relative size of the bar.values |
cex.names |
character extension or relative scale of the name labels for the bars |
main |
main title of the graph |
xlab |
label of X axis |
ylab |
label of Y axis |
col |
colours of the bar |
gen.ind.vars |
whether the indicator variables will be generated |
x |
object of class 'tab1' obtained from saving 'tab1' results |
... |
further arguments passed to or used by other methods |
'tab1' is an advanced one-way tabulation providing a nice frequency table as well as a bar chart. The description of the variable is also used in the main title of the graph.
The bar chart is vertical unless the number of categories is more than six and any of the labels of the levels consists of more than 8 characters or 'horiz' is set to TRUE.
For table has less than categories, the automatic colour is "grey". Otherwise, the graph will be colourful. The argument, 'col' can be overwritten by the user.
The argument 'gen.ind.vars' is effective only if x0 is factor.
Output table
Virasakdi Chongsuvivatwong [email protected]
'tabpct', 'label.var', 'table', 'barplot', 'model.matrix'
tab1(state.division) tab1(state.division, bar.values ="percent") tab1(state.division, sort.group ="decreasing") tab1(state.division, sort.group ="increasing") tab1(state.division, col=c("chocolate","brown1","brown4"), main="Number of states in each zone") # For presentation, several 'cex' parameters should increase tab1(state.division, col=c("chocolate","brown1","brown4"), main="Number of states in each zone", cex.main=1.7, cex.name=1.2, cex.axis=1.3, cex.lab=1.3) data(Oswego) .data <- Oswego attach(.data) tab1(ill) # Note the column of cumulative percentages in the table. tab1(ill, cum.percent=FALSE) tab1(chocolate) # Due to missing values, cumulative percentages are now automatically turned off. tab1(chocolate, cum.percent=TRUE) # Slightly too many columns in text! tab1(chocolate, missing=FALSE, bar.values="percent") agegr <- cut(age, breaks=c(0,10,20,30,40,50,60,70,80)) tab1(agegr) # No need to start with 'calc' as it is outside .data tab1(agegr, col="grey") # graphic output from older versions of 'tab1' tab1(agegr, col=c("red","yellow","blue")) # Colours recycled tab1(agegr, horiz=TRUE) # Keeping output table dev.off() tab1(agegr, graph = FALSE) -> a print(a) a # same results attributes(a) a$output.table class(a$output.table) # "matrix" # 'a$output.table' is ready for exporting to a .csv file by # write.csv(a$output.table, file="table1.csv") # "table1.csv" is now readable by a spreadsheet program detach(.data) rm(list=ls())tab1(state.division) tab1(state.division, bar.values ="percent") tab1(state.division, sort.group ="decreasing") tab1(state.division, sort.group ="increasing") tab1(state.division, col=c("chocolate","brown1","brown4"), main="Number of states in each zone") # For presentation, several 'cex' parameters should increase tab1(state.division, col=c("chocolate","brown1","brown4"), main="Number of states in each zone", cex.main=1.7, cex.name=1.2, cex.axis=1.3, cex.lab=1.3) data(Oswego) .data <- Oswego attach(.data) tab1(ill) # Note the column of cumulative percentages in the table. tab1(ill, cum.percent=FALSE) tab1(chocolate) # Due to missing values, cumulative percentages are now automatically turned off. tab1(chocolate, cum.percent=TRUE) # Slightly too many columns in text! tab1(chocolate, missing=FALSE, bar.values="percent") agegr <- cut(age, breaks=c(0,10,20,30,40,50,60,70,80)) tab1(agegr) # No need to start with 'calc' as it is outside .data tab1(agegr, col="grey") # graphic output from older versions of 'tab1' tab1(agegr, col=c("red","yellow","blue")) # Colours recycled tab1(agegr, horiz=TRUE) # Keeping output table dev.off() tab1(agegr, graph = FALSE) -> a print(a) a # same results attributes(a) a$output.table class(a$output.table) # "matrix" # 'a$output.table' is ready for exporting to a .csv file by # write.csv(a$output.table, file="table1.csv") # "table1.csv" is now readable by a spreadsheet program detach(.data) rm(list=ls())
Tabulation of variables with the same possible range of distribution and stack into a new table with or without other descriptive statistics or to breakdown distribution of more than one row variables against a column variable
tableStack (vars, dataFrame, minlevel = "auto", maxlevel = "auto", count = TRUE, na.rm =FALSE, means = TRUE, medians = FALSE, sds = TRUE, decimal = 1, total = TRUE, var.labels = TRUE, var.labels.trunc =150, reverse = FALSE, vars.to.reverse = NULL, by = NULL, vars.to.factor = NULL, iqr = "auto", prevalence = FALSE, percent = "column", frequency=TRUE, test = TRUE, name.test = TRUE, total.column = FALSE, simulate.p.value = FALSE, sample.size=TRUE, assumption.p.value = .01)tableStack (vars, dataFrame, minlevel = "auto", maxlevel = "auto", count = TRUE, na.rm =FALSE, means = TRUE, medians = FALSE, sds = TRUE, decimal = 1, total = TRUE, var.labels = TRUE, var.labels.trunc =150, reverse = FALSE, vars.to.reverse = NULL, by = NULL, vars.to.factor = NULL, iqr = "auto", prevalence = FALSE, percent = "column", frequency=TRUE, test = TRUE, name.test = TRUE, total.column = FALSE, simulate.p.value = FALSE, sample.size=TRUE, assumption.p.value = .01)
vars |
a vector of variables in the data frame |
dataFrame |
source data frame of the variables |
minlevel |
possible minimum value of items specified by user |
maxlevel |
possible maximum value of items specified by user |
count |
whether number of valid records for each item will be displayed |
na.rm |
whether missing value would be removed during calculation mean score of each person |
means |
whether means of all selected items will be displayed |
medians |
whether medians of all selected items will be displayed |
sds |
whether standard deviations of all selected items will be displayed |
decimal |
number of decimals displayed in the statistics |
total |
display of means and standard deviations of total and average scores |
var.labels |
presence of descriptions of variables on the last column of output |
var.labels.trunc |
number of characters used for variable description |
reverse |
whether item(s) negatively correlated with other majority will be reversed |
vars.to.reverse |
variable(s) to reverse |
by |
a variable for column breakdown. If a single character (with quotes) is given, only the 'total column' will be displayed |
vars.to.factor |
variable(s) to be converted to factor for tabulaton |
iqr |
variable(s) to display median and inter-quartile range |
prevalence |
for logical variable, whether prevalence of the dichotomous row variable in each column subgroup will be displayed |
percent |
type of percentage displayed when the variable is categorical. Default is "column". It can be "row", or "none" |
frequency |
whether to display frequency in the cells when the variable is categorical |
test |
whether statistical test(s) will be computed |
name.test |
display name of the test and relevant degrees of freedom |
total.column |
whether to add 'total column' to the output or not |
simulate.p.value |
simulate P value for Fisher's exact test |
sample.size |
whether to display non-missing sample size of each column |
assumption.p.value |
level of Bartlett's test P value to judge whether the comparison and the test should be parametric |
This function simultaneously explores several variables with a fixed integer rating scale. For non-factor variables, the default values for tabulation are the minimum and the maximum of all variables but can be specified by the user.
When 'by' is omitted, all variables must be of the same class, and must be 'integer', 'factor' or 'logical.
Unlike function 'alpha', the argument 'reverse' has a default value of FALSE. This argument is ignored if 'vars.to.reverse' is specified.
Options for 'reverse', 'vars.to.reverse' and statistics of 'means', 'medians', 'sds' and 'total' are available only if the items are not factor. To obtain statistics of factor items, users need to use 'unclassDataframe' to convert them into integer.
When the 'by' argument is given, 'reverse' and 'vars.to.reverse' do not apply. Instead, columns of the 'by' variable will be formed. A table will be created against each selected variable. If the variable is a factor or coerced to factor with 'vars.to.factor', cross-tabulation will result with percents as specified, ie. "column", "row", or "none" (FALSE). For a dichotomous row variable, if set to 'TRUE', the prevalence of row variable in the form of a fraction is displayed in each subgroup column. For objects of class 'numeric' or 'integer', means with standard deviations will be displayed. For variables with residuals that are not normally distributed or where the variance of subgroups are significantly not normally distributed (using a significance level of 0.01), medians and inter-quartile ranges will be presented if the argument 'iqr' is set to "auto" (by default). Users may specify a subset of the selected variables (from the 'vars' argument) to be presented in such a form. Otherwise, the argument could be set as any other character string such as "none", to insist to present means and standard deviations.
When 'test = TRUE' (default), Pearson's chi-squared test (or a two-sided Fisher's exact test, if the sample size is small) will be carried out for a categorical variable or a factor. Parametric or non-parametric comparison and test will be carried out for a object of class 'numeric' or 'integer' (See 'iqr' and 'assumption.p.value' below). If the sample size of the numeric variable is too small in any group, the test is omitted and the problem reported.
For Fisher's exact test, the default method employs 'simulate.p.value = FALSE'. See further explanation in 'fisher.test' procedure. If the dataset is extraordinarily large, the option may be manually set to TRUE.
When 'by' is specified as a single character object (such as 'by="none"'), there will be no column breakdown and all tests will be omitted. Only the total column is displayed. Only the 'total' column is shown.
If this 'total column' is to accompany the 'by' breakdown, the argument 'total.column=TRUE' should be specified. The 'sample.size' is TRUE by default. The total number of records for each group is displayed in the first row of the output. However, the variable in each row may have some missing records, the information on which is not reported by tableStack.
By default, Epicalc sets 'var.labels=TRUE' in order to give nice output. However, 'var.labels=FALSE' can sometimes be more useful during data exploration. Variable numbers as well as variable names are displayed instead of variable labels. Names and numbers of abnormally distributed variables, especially factors with too many levels, can be easily identified for further relevelling or recoding.
The argument 'iqr' has a default value being "auto". Non-parametric comparison and test will be automatically chosen if Bartlett's test P value is below the 'assumption.p.value'.
The test can be forced to parametric by setting 'iqr=NULL' and to non-parametric by if iqr is set to the variable number of cont.var (See examples.).
an object of class 'tableStack' and 'list' when by=NULL
results |
an object of class 'noquote' which is used for print out |
items.reversed |
name(s) of variable(s) reversed |
total.score |
a vector from 'rowSums' of the columns of variables specified in 'vars' |
mean.score |
a vector from 'rowMeans' of the columns of variables specified in 'vars' |
mean.of.total.scores |
mean of total scores |
sd.of.total.scores |
standard deviation of total scores |
mean.of.average.scores |
mean of mean scores |
sd.of.average.scores |
standard deviation of mean scores |
When 'by' is specified, an object of class 'tableStack' and 'table is returned.
Virasakdi Chongsuvivatwong [email protected]
'table', 'tab1', 'summ', 'alpha', 'unclassDataframe'
data(Oswego) tableStack(bakedham:fruitsalad, dataFrame=Oswego) .data <- Oswego des(.data) attach(.data) tableStack(bakedham:fruitsalad, .data) # Default data frame is .data tableStack(bakedham:fruitsalad, .data, by= ill) tableStack(bakedham:fruitsalad, .data, by= ill, prevalence=TRUE) tableStack(bakedham:fruitsalad, .data, by= ill, percent=FALSE) tableStack(bakedham:fruitsalad, .data, by= ill, percent=FALSE, name.test=FALSE) detach(.data) data(Cars93, package="MASS") .data <- Cars93 des(.data) tableStack(vars=4:25, .data, by=Origin) tableStack(vars=4:25, .data, by="none") tableStack(vars=4:25, .data, by=Origin, total.column=TRUE) data(Attitudes) .data <- Attitudes attach(.data) tableStack(qa1:qa18, .data) # May need full screen of Rconsole tableStack(qa1:qa18, .data, var.labels.trunc=35) # Fits in with default R console screen tableStack(qa1:qa18, .data, reverse=TRUE) -> a a ## Components of 'a' have appropriate items reversed a$mean.score -> mean.score a$total.score -> total.score .data$mean.score <- mean.score .data$total.score <- total.score rm(total.score, mean.score) detach(.data) attach(.data) tableStack(c(qa1,qa13:qa18,mean.score,total.score), .data, by=sex, test=FALSE) tableStack(c(qa15, qa17, mean.score:total.score), .data, by=sex, iqr=c(qa17,total.score)) tableStack(c(qa15, qa17, mean.score:total.score), .data, by=dep, iqr=c(qa17,total.score)) ## 'vars' can be mixture of different classes of variables .data$highscore <- mean.score > 4 tableStack(mean.score:highscore, .data, by=sex, iqr=total.score) detach(.data) rm(list=ls()) data(Ectopic) .data <- Ectopic des(.data) tableStack(vars=3:4, .data, by=outc) tableStack(vars=3:4, .data, by=outc, percent="none") tableStack(vars=3:4, .data, by=outc, prevalence = TRUE) tableStack(vars=3:4, .data, by=outc, name.test = FALSE) ## Variable in numeric or factor data(Outbreak) .data <- Outbreak des(.data) # Comparison of exposure to food items between the two gender tableStack(vars=5:8, .data, by=sex) # as continuous varaibles tableStack(vars=5:8, .data, by=sex, vars.to.factor = 5:8) # as factorsdata(Oswego) tableStack(bakedham:fruitsalad, dataFrame=Oswego) .data <- Oswego des(.data) attach(.data) tableStack(bakedham:fruitsalad, .data) # Default data frame is .data tableStack(bakedham:fruitsalad, .data, by= ill) tableStack(bakedham:fruitsalad, .data, by= ill, prevalence=TRUE) tableStack(bakedham:fruitsalad, .data, by= ill, percent=FALSE) tableStack(bakedham:fruitsalad, .data, by= ill, percent=FALSE, name.test=FALSE) detach(.data) data(Cars93, package="MASS") .data <- Cars93 des(.data) tableStack(vars=4:25, .data, by=Origin) tableStack(vars=4:25, .data, by="none") tableStack(vars=4:25, .data, by=Origin, total.column=TRUE) data(Attitudes) .data <- Attitudes attach(.data) tableStack(qa1:qa18, .data) # May need full screen of Rconsole tableStack(qa1:qa18, .data, var.labels.trunc=35) # Fits in with default R console screen tableStack(qa1:qa18, .data, reverse=TRUE) -> a a ## Components of 'a' have appropriate items reversed a$mean.score -> mean.score a$total.score -> total.score .data$mean.score <- mean.score .data$total.score <- total.score rm(total.score, mean.score) detach(.data) attach(.data) tableStack(c(qa1,qa13:qa18,mean.score,total.score), .data, by=sex, test=FALSE) tableStack(c(qa15, qa17, mean.score:total.score), .data, by=sex, iqr=c(qa17,total.score)) tableStack(c(qa15, qa17, mean.score:total.score), .data, by=dep, iqr=c(qa17,total.score)) ## 'vars' can be mixture of different classes of variables .data$highscore <- mean.score > 4 tableStack(mean.score:highscore, .data, by=sex, iqr=total.score) detach(.data) rm(list=ls()) data(Ectopic) .data <- Ectopic des(.data) tableStack(vars=3:4, .data, by=outc) tableStack(vars=3:4, .data, by=outc, percent="none") tableStack(vars=3:4, .data, by=outc, prevalence = TRUE) tableStack(vars=3:4, .data, by=outc, name.test = FALSE) ## Variable in numeric or factor data(Outbreak) .data <- Outbreak des(.data) # Comparison of exposure to food items between the two gender tableStack(vars=5:8, .data, by=sex) # as continuous varaibles tableStack(vars=5:8, .data, by=sex, vars.to.factor = 5:8) # as factors
Two-way tabulation with automatic mosaic plot
tabpct(row, column, decimal = 1, percent = "both", graph = TRUE, las = 0, main = "auto", xlab = "auto", ylab = "auto", col = "auto", ...)tabpct(row, column, decimal = 1, percent = "both", graph = TRUE, las = 0, main = "auto", xlab = "auto", ylab = "auto", col = "auto", ...)
row, column
|
variables |
decimal |
number of decimals for the percentage in the table |
percent |
orientation of the percentage in the table |
graph |
automatic graphing |
las |
orientation of group labelling |
main |
main title |
xlab |
X axis label |
ylab |
Y axis label |
col |
colours of the bars |
... |
additional arguments for 'table' |
0: always parallel to axis
1: always horizontal,
2: always perpendicular to the axis,
3: always vertical.
'tabpct' gives column and row percent cross-tabulation as well as mosaic plot.
The width of the bar in the plot denotes the relative proportion of the row variable.
Inside each bar, the relative proportion denotes the distribution of column variables within each row variable.
The default value for the 'percent' orientation is "both". It can also be "col" or "row".
Due to limitation of 'mosaicplot', certain graphic parameters such as 'cex.main', 'cex.lab' are not acceptable. The parameter 'main', 'xlab' and 'ylab' can be suppressed by making equal to " ". An additional line starting with 'title' can be used to write new main and label titles with 'cex.main' and 'cex.lab' specified.
Tables of row and column percentage
Virasakdi Chongsuvivatwong [email protected]
'tab1', 'table', 'mosaicplot'
data(Oswego) .data <- Oswego attach(.data) agegr <- cut(age, breaks=c(0,20,40,60,80)) tabpct(agegr, ill) tabpct(agegr, ill, cex.axis=1) # enlarge value labels # To increase the size of the various titles: tabpct(agegr, ill, cex.axis=1, main="", xlab="", ylab="", col=c("blue","purple")) title(main="Diseased by Age group", cex.main=1.8, xlab="Age (years)",ylab="Diseased", cex.lab=1.5) detach(.data) rm(list=ls())data(Oswego) .data <- Oswego attach(.data) agegr <- cut(age, breaks=c(0,20,40,60,80)) tabpct(agegr, ill) tabpct(agegr, ill, cex.axis=1) # enlarge value labels # To increase the size of the various titles: tabpct(agegr, ill, cex.axis=1, main="", xlab="", ylab="", col=c("blue","purple")) title(main="Diseased by Age group", cex.main=1.8, xlab="Age (years)",ylab="Diseased", cex.lab=1.5) detach(.data) rm(list=ls())
This dataset came from an interview survey on the workshop attendants on 2004-12-14.
Note that the date of bed time is 2004-12-13 if the respondent went to bed before midnight.
data(Timing)data(Timing)
A data frame with 18 observations on the following 11 variables.
ida numeric vector
gendera factor with levels male female
agea numeric vector
maritala factor with levels single married others
childa numeric vector indicating number of children
bedhra numeric vector indicating the hour of going to bed
bedmina numeric vector indicating the minute of going to bed
wokhra numeric vector indicating the hour of waking up
wokmina numeric vector indicating the minute of waking up
arrhra numeric vector indicating the hour of arrival at the workshop
arrmina numeric vector indicating the minute of arrival at the workshop
data(Timing) des(Timing)data(Timing) des(Timing)
Setting vocabularies for Epicalc graph title
titleString (distribution.of = .distribution.of, by = .by, frequency = .frequency, locale = .locale(), return.look.up.table=FALSE)titleString (distribution.of = .distribution.of, by = .by, frequency = .frequency, locale = .locale(), return.look.up.table=FALSE)
distribution.of |
A string denoting "Distribution of" |
by |
That for "by" |
frequency |
That for "Frequency" |
locale |
Logical value to overwrite .locale(). The initial value is FALSE |
return.look.up.table |
Should the look-up table be returned? |
The two internationalization commands of Epicalc, 'setTitle' and 'titleString', work together to set the langauge and wording of titles of automatic graphs obtained from certain Epicalc functions.
In general, 'setTitle' is simple and works well if the locale required fits in with the version of the operating system. The three commonly used words in the graph titles: "Distribution of", "by" and "Frequency", which are in English, are initially stored in three respective hidden objects '.distribution.of', '.by' and '.frequency' as well as in the look-up table within the 'titleString' function. When the locale is changed to a language other than English, the look-up table is used and wordings are changed accordingly.
The function 'titleString' is useful when the user wants to change the strings stored in the look-up table. It changes the initial values of '.distribution.of', '.by' and '.frequency', respectively. The argument, 'locale', must be manually set to FALSE by the user to disable the use of the look-up table and to enable the use of the three objects assigned by the command instead.
The two functions suppress each other. Use of 'setTitle' disables the effects of 'titleString', switching .locale() to TRUE and forcing Epicalc to read from the look-up table in 'titleString'. However, 'setTitle' does not overwrite the values assigned by the arguments of 'titleString'.
The key and decisive switch object is .locale(). Once .locale() is set to FALSE, either manually or inside the 'titleString' command, the values of the three hidden objects will be used. Setting .locale() to TRUE, either manually or automatically by the 'setTitle' function, points the graph title to use the look-up table inside 'titleString'.
Typing 'titleString()' without an argument displays the current contents of these three objects. The look-up table is also displayed if the return.look.up.table argument is set to TRUE.
International users who want to add their specific locales and corresponding terminology to the look-up table or to suggest more appropriate terminology can contact the author.
Virasakdi Chongsuvivatwong [email protected]
'setTitle'
.data <- iris attach(.data) dotplot(Sepal.Length, by=Species) titleString(distribution.of="", by="grouped by", locale=FALSE) ## The above command is equivalent to the following three lines: ## .distribution.of <- "" ## .by <- "grouped by" ## .locale(FALSE) dotplot(Sepal.Length, by=Species) titleString() setTitle("English") dotplot(Sepal.Length, by=Species) titleString(return.look.up.table=TRUE) .locale(FALSE) dotplot(Sepal.Length, by=Species) titleString() .distribution.of <- "Distribution of" titleString() .by <- "by" titleString() detach(.data) rm(.data).data <- iris attach(.data) dotplot(Sepal.Length, by=Species) titleString(distribution.of="", by="grouped by", locale=FALSE) ## The above command is equivalent to the following three lines: ## .distribution.of <- "" ## .by <- "grouped by" ## .locale(FALSE) dotplot(Sepal.Length, by=Species) titleString() setTitle("English") dotplot(Sepal.Length, by=Species) titleString(return.look.up.table=TRUE) .locale(FALSE) dotplot(Sepal.Length, by=Species) titleString() .distribution.of <- "Distribution of" titleString() .by <- "by" titleString() detach(.data) rm(.data)
Relationship between bacteria and presence of any decayed tooth.
data(Decay)data(Decay)
A data frame with 436 observations on the following 2 variables.
decaya numeric vector indicating presence of tooth decay
strepa numeric vector indicating number of colony-forming-units (CFUs) of Streptococcus mutan in the saliva
Teanpaisan, R., Kintarak, S., Chuncharoen, C., Akkayanont, P. 1995 Mutans Streptococci and dental -caries in schoolchildren in Southern Thailand. Community Dentistry and Oral Epidemiology 23: 317-318.
data(Decay) des(Decay)data(Decay) des(Decay)
This dataset contains information on the records of 200 women working at a tourist destination community.
data(VCT)data(VCT)
A subset of a data frame containing 200 observations and 12 variables with variable descriptions.
Details of the codes can be seen from the results of the function 'codebook()' below.
data(VCT) codebook(VCT)data(VCT) codebook(VCT)
This dataset was adopted from Diggle et al: Analysis of Longitudinal Data. REFERENCE – Zeger and Karim, JASA (1991)
Note that there are some duplications of id and time combination.
data(Xerop)data(Xerop)
A data frame containing 1200 observations and 10 variables.
ida numeric vector for personal identification number
respinfectwhether the child had respiratory infection in that visit
age.monthcurrent age in month
xeropwhether the child currently had vitamin A deficiency
sexgender of the child no detail on the code
ht.for.ageheight for age
stuntedwhether the child has stunted growth
timetime of scheduled visit
baseline.agebaseline age
seasonseason
data(Xerop)data(Xerop)