Effective graphs for data display

Introduction
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12

Introduction

This page contains Stata code and output, equivalent to the examples in the paper. In all cases, the data are random samples from the freely-available NHANES dataset, and the variables used are;

BPXSAR: systolic blood pressure (mmHg)
BPXDAR: diastolic blood pressure (mmHg)
BPXDI1, BPXDI2: two diastolic blood pressure readings
race_ethc: race/ethnicity, coded as Hispanic, White non-Hispanic, Black non-Hispanic and Other
DR1TFOLA: folate intake (\(\mu\)g/day)
RIAGENDR: sex, coded as Male/Female
BMXBMI: body mass index (kg/m\(^2\))
RIDAGEY: age (years)

To run the code, the following packages must first be installed;

vioplot: ssc install vioplot
stripplot: ssc install stripplot
spineplot: ssc install spineplot

Following this, cut-and-paste the code for each figure into an R session.

Figure 1

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhanessmall.csv, clear
label variable bpxsar "Systolic Blood Pressure (mmHg)"
stripplot bpxsar

Figure 1a

stripplot bpxsar, stack center

Figure 1b

Figure 2

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhanesmedium.csv, clear
label variable bpxsar "Systolic Blood Pressure (mmHg)"
stripplot bpxsar

Figure 2a

stripplot bpxsar, stack center

Figure 2b

stripplot bpxsar, width(1) stack center

Figure 2c

Figure 3

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
label variable bpxsar "Systolic Blood Pressure (mmHg)"
vioplot bpxsar, horizontal

Figure 3a

histogram bpxsar

Figure 3b

Figure 4

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
generate normotens = bpxsar <= 140
label define normotensl 1 "<= 140 mmHg" 0 "> 140 mmHg" 
label values normotens normotensl
generate sample = 1
graph hbar (percent) sample, over(normotens) asyvars stack

Figure 4a

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
generate normotens = bpxsar <= 140
collapse (mean) est=normotens (sebinomial) stderr=normotens
generate loCI = est - 1.96*stderr
generate hiCI = est + 1.96*stderr
generate x = 1
label variable x "Estimate"
label variable loCI "95% Conf Int"
label variable hiCI "95% Conf Int"
twoway (scatter x est) (rspike loCI hiCI x, horizontal), ///
xscale( r(0 1) ) xtick( 0(0.2)1 ) xlabel( 0(0.2)1) yscale(off)

Figure 4b

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
generate sample = 1
graph hbar (percent) sample, over(race_ethc) asyvars stack

Figure 4c

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
generate x=1
collapse (count) x, by(race_ethc)
generate est = x/1000
generate stderr = sqrt(est*(1-est)/1000)
generate loCI = est - 1.96*stderr
generate hiCI = est + 1.96*stderr
generate study = _n
label variable study "Estimate"
label variable loCI "95% Conf Int"
label variable hiCI "95% Conf Int"
twoway (scatter study est) (rspike loCI hiCI study, horizontal), ///
xscale( r(0 1) ) xtick( 0(0.2)1 ) xlabel( 0(0.2)1) /// 
ylabel( 1 "Black non-Hispanic" 2 "Hispanic" 3 "Other" 4 "White non-Hispanic", angle(0)) ///
ytitle("")

Figure 4d

Figure 5

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
generate normotens = bpxsar <= 140
label define normotensl 1 "<= 140 mmHg" 0 "> 140 mmHg" 
label values normotens normotensl
generate x = 1
graph pie x, over(normotens)

Figure 5a

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhaneslarge.csv, clear
generate x = 1
graph pie x, over(race_ethc)

Figure 5b

Figure 6

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhanessmall.csv, clear
label variable dr1tfola "Folate intake (ug/day)"
label define sexl 1 "Male" 2 "Female" 
label values riagendr sexl
stripplot dr1tfola, vertical over(riagendr)

Figure 6a

insheet using http://faculty.washington.edu/kenrice/heartgraphs/nhanesmedium.csv, clear
label variable dr1tfola "Folate intake (ug/day)"
label define sexl 1 "Male" 2 "Female" 
label values riagendr sexl
stripplot dr1tfola, width(20) vertical stack center over(riagendr)