#Day30| Uncertainties – Data Day – Global Health Data Exchange | #30DayChartChallenge | Life Expectancy at Birth — Latin America & Caribbean. Built with #RStats using #ggplot2, #patchwork, #scales, #grid, #gridExtra and #tidyr.
#Day29 | Uncertainties – Monochrome | #30DayChartChallenge | . Coffee Price Forecast — Holt-Winters (HW) Built with #RStats using #forecast, #ggplot2, #dplyr, #lubridate, #scales and #tidyr.
#Day9 | Distributions – Wealth | #30DayChartChallenge | Income Distribution in Central America, source World Bank. Built with #RStats using #ggplot2, #dplyr, #tidyr, #patchwork, #ggtext, #scales, #wbstats and #purrr.
#Día5 | Comparaciones – Experimental | #30DayChartChallenge. Experimenté agregando una sumatoria horizontal de observaciones en un boxplot sobre la capacidad endocraneana en especies del género Homo. Creada usando R con #ggplot2, #ggdist, #dplyr, #scales, #ggtext, #patchwork, #tibble y #tidyr.
#Día 2 | Comparaciones – Pictograma | #30DayChartChallenge. Centroamérica suma más de 51 millones de habitantes. El gráfico fue creada usando R con #ggplot2, #dplyr, #tidyr#, #scales, #ggflags, #sf, #rnaturalearth, #rnaturalearthdata, #patchwork.

Finally sat down to try to get my head around `reshape()` after being told somewhere that `melt`/`cast` are old hat, `reshape` is the new hotness...the result?

Friendship ended with `reshape`, now me and #tidyr are besties, `pivot_longer`/`pivot_wider` ilu  #rlang #dataviz

BTW, so far I have not encountered any scenario in which #tidyR offers solutions superior to #baseR.

I can't speak for anyone else, but in my line of work, I achieve everything I want to do in base R with fewer lines of code than with what tidyR, dplyr and the like have to offer.

It's a 72nd day of my #100DaysOfCoding with #R and #RStats. I met #tidyr and #purrr.

@edyhsgr Potentially controversial, but I agree!

I was annoyed when the code demos for a particular uni course relied on #tidyR, because I felt that I hadn't learned enough #baseR yet.

So I stuck to base for my big assignment, only bringing in other packages where needed. It's not that I dislike tidy; it's just that I want to be in control of the packages - I don't want the packages to be in control of me.

Any #rstats users here that use #tidyr et al. for clustering and ordination?

I'm trying to wrap my head around how I should manage a workflow like:

data -> distance matrix -> clustering or ordination

*without* the benefit of row names to link the original data to the resulting cluster leaf or ordination point.