Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Based on some feedback to a previous post I normalised the birth counts by the (average) number of days in each month. As pointed out by a reader, the results indicate a gradual increase in the number of conceptions during (northern hemisphere) Autumn and Winter, roughly up to the end of December. Normalising the data to give births per day also shifts the peak from August to September.

> library(lifespan)
> month.days <- data.frame(
+   month = month.abb,
+   days = c(31, 28.25, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31)
+ )
> group_by(births, month) %>% summarise(count = sum(count)) %>%
+   merge(month.days) %>%
+   mutate(perday = count / days) %>%
+   ggplot(aes(x = month, y = perday / 1000)) +
+   geom_bar(stat = "identity", fill = "#39A75E") +
+   labs(x = "", y = "Total Births per Day [thousands]") +
+   theme_classic()


I also broke the births data down by gender. The August peak persists for both genders but female births exhibit consistently higher variability that male births. For that I have no plausible explanation. Suggestions?

> group_by(births, year, month, sex) %>% summarise(count = sum(count)) %>%
+   merge(month.days) %>%
+   mutate(perday = count / days) %>% ggplot() +
+   geom_boxplot(aes(x = month, y = perday, fill = sex)) +
+   labs(x = "", y = "Births per Day") +
+   theme_classic() + theme(legend.title = element_blank())