Share of Symptoms graphs

This document has compiled all of the graphs that observe the share of symptoms that follow some variation of precursor subdivision. The timeframe that was used in these particular visualizations is 6 months in that symptoms were counted as a follwer of a precursor if they followed that precursor within 6 months excluding the same month that the precursor occured in. Each of these graphs are additionally disaggregated by regime type, democracy or autocracy, in order to better understand how these regime types effect our observations.

## Rows: 23421 Columns: 20
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (15): Country, ISO, Type, Category, Month_1, Month_2, Actors, Targets, E...
## dbl  (5): FH_year, Year_1, Year_2, v2x_regime, e_regionpol_6C
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

## Warning in inner_join(., v7 %>% filter(Type == "Symptom"), by = "Country", : Detected an unexpected many-to-many relationship between `x` and `y`.
## ℹ Row 1 of `x` matches multiple rows in `y`.
## ℹ Row 1 of `y` matches multiple rows in `x`.
## ℹ If a many-to-many relationship is expected, set `relationship =
##   "many-to-many"` to silence this warning.

Before diving right into these visualization, it’s first important to discuss what is being shown in the graphs. The share of symptoms is just a percent of the total symptoms that are observed in the dataset. For each graph I will give a specific interpretation to help guide you in your understanding of this data.

By looking at the longest bar in the Share of Symptoms Following a Precursor that is Horizontal (Democracies) graph we can see that it falls at 50%. What this means is that 50% of all external shock symptoms that are observed in democracies come within 6 months of a precursor. While I don’t believe that the longest bar is the most interesting observation, I think that the second and third longest bars are a place that could be considered for further study. A sizeable portion of those symptoms are preceded by precursors and better understanding what those precursors are can help us better identitfy when democracies are at risk of experiencing electoral violence or regime change.

It seems that in authoritarian regimes, symptoms are not as often followed by precursors which makes sense for the nature of these regimes. There is no need to erode certain legal boundaries to infringe on people’s rights or no need to gradually accustom citizens to mistreatment, they can sort of immediately jump to exhibiting symptoms.

It’s interesting here as well that external shocks jumps out at the very front in democracies yet again. Ignoring that outlier however, it seems that vertical precursors are not as predictive of ongoing democratic backsliding in democracies as the horizontal precursors.

In authoritarian states it seems that vertical precursors are moderately predictive in whether or not the nation experiences certain symptoms like politicization of the education system or foreign militaryh action.

Both of the exogenous graphs I feel tell us very little about both regime types and I think that is in part due to the nature of these precursors. Exogenous precursors are not exactly controlled by states. Natural disasters happen and pandemics suddenly shut down the world. These events are not exactly state-controlled

Both of these graphs are just looking at if symptoms generally follow a precursor. While it’s less informative than the binned version of the same graph (split between horizontal, vertical, and exogenous precursors), at least in the democracy graph, precursors are pretty important to predicting a great deal of backsliding symptoms.