New Figure Schemes for Stata: blindschemes The Schemes plotplain - PowerPoint PPT Presentation

The Issue Solutions New Figure Schemes for Stata: blindschemes The Schemes plotplain & plottig Adaptation Conclusion Daniel Bischof 1 1 University of Zurich November 17, 2016 1/17

The Default Stata Figure Schemes 6 The Issue Solutions 5 blindschemes The Schemes 4 Adaptation Conclusion 3 2 1 4 5 6 7 logged number of terrorist events 95% CI Fitted values logged number of killed people 2/17

Limitations The Issue Solutions colors are difficult to differentiate for colorblind people blindschemes The Schemes background tinting Adaptation frames Conclusion symbols, markers, lines often too thick gridlines only parallel to x-axis legends could be placed closer to content of figure 3/17

Solutions The Issue Solutions blindschemes The Schemes Adaptation 1 the obvious solution is to produce code Conclusion 4/17

Solutions The Issue Solutions blindschemes The Schemes Adaptation 1 the obvious solution is to produce code; lots of code . . . Conclusion 4/17

Solutions The Issue twoway /// Solutions (line numcountries year, lcolor(gs12)) /// blindschemes , ylabel(, angle(horizontal)) xtitle("") /// The Schemes Adaptation graphregion(fcolor(white) lcolor(white) Conclusion lwidth(vvvthick) ifcolor(white) ilcolor(white) ilwidth(vvvthick)) /// plotregion(lcolor(white) lwidth(vvvthick) ifcolor(white) ilcolor(white) ilwidth(vvvthick)) /// legend(cols(1) region(lcolor(white))) 5/17

Solutions The Issue Solutions blindschemes 1 the obvious solution is to produce code; lot’s of code . . . The Schemes Adaptation 2 use Billy Buchanan’s brewscheme to define your own Conclusion designs 6/17

Solutions The Issue Solutions blindschemes 1 the obvious solution is to produce code; lot’s of code . . . The Schemes Adaptation 2 use Billy Buchanan’s brewscheme to define your own Conclusion designs 3 write a new package addressing some of the key limitations 6/17

Why write a new figure scheme? The Issue Solutions blindschemes ! time The Schemes ! ensure quality (create uniform norm) Adaptation ! simplify the usage of more and new colors Conclusion ! many users lack knowledge how to adapt figures ⇒ there seems to be a high demand for new and alternative figure schemes 7/17

Solution 3: Write a new figure scheme The Issue Solutions I wrote 2 figure schemes: blindschemes The Schemes plotplain: very simple, “clean” figure scheme Adaptation plottig: replicates ggplot2 (R) by Hadley Wickham in most Conclusion regards 8/17

Solution 3: Write a new figure scheme The Issue Solutions I wrote 2 figure schemes: blindschemes The Schemes plotplain: very simple, “clean” figure scheme Adaptation plottig: replicates ggplot2 (R) by Hadley Wickham in most Conclusion regards → both available with colors distinguishable for colorblind people 8/17

Solution 3: blindschemes in the SSC Archive The Issue Solutions blindschemes The Schemes Adaptation Conclusion 9/17

Plotplain The Issue Solutions 6 blindschemes 5 The Schemes Adaptation 4 Conclusion 3 2 Western Europe North America MENA 1 4 5 6 7 logged number of terrorist events 10/17

Plottig The Issue Solutions 6 blindschemes 5 The Schemes Adaptation 4 Conclusion 3 2 Western Europe North America MENA 1 4 5 6 7 logged number of terrorist events 11/17

The colorblind colors plotplainblind 10 black The Issue gs10 8 Solutions sky turquoise blindschemes 6 orangebrown The Schemes reddish 4 vermillion Adaptation sea 2 Conclusion ananas 0 plottig 15 black plr1 ply1 plg1 plb1 10 pll1 plr2 ply2 plg2 plb2 5 pll2 ply3 plg3 plb3 pll3 0 12/17

Adaptation of Code: Gridlines The Issue Solutions blindschemes The Schemes lpoly nkill iyear if region==12, /// Adaptation ci legend(off) title("Polynomial fit of terrorist Conclusion casualties in Western Europe") /// note("") 13/17

Adaptation of Code: Gridlines Polynomial fit of terrorist casualties in Western Europe The Issue 6 Solutions blindschemes 5 The Schemes logged number of killed people Adaptation 4 Conclusion 3 2 1 1970 1980 1990 2000 2010 2020 incident year 14/17

Adaptation of Code: Gridlines The Issue Solutions blindschemes The Schemes lpoly nkill iyear if region==12, /// Adaptation ci legend(off) title("Polynomial fit of terrorist Conclusion casualties in Western Europe") /// note("") xlabel(, nogrid) 15/17

Adaptation of Code: Gridlines Polynomial fit of terrorist casualties in Western Europe The Issue 6 Solutions blindschemes 5 The Schemes logged number of killed people Adaptation 4 Conclusion 3 2 1 1970 1980 1990 2000 2010 2020 incident year 16/17

Conclusion The Issue + Improvement: Less code needed, users can focus on other Solutions tasks blindschemes The Schemes + Disagreement: Even if you disagree, less changes are Adaptation needed for further adaptation Conclusion - Concerns: Still issues remain, e.g. overlapping confidence intervals For further information including the working paper on the schemes: danbischof.com 17/17