Background and goal

After doing 3 rounds of analysis we were able to identify coalitions more clearly and produce a 3-period explanation. It was a process that forced to verify multiple things, including the coding of certain statements.

In this file, we will depart from the data imported to visone to:

verify the number of organizations involved in the debate (824 orgs according to DNA) and
calculate the segmentation of the total sample by organization type

These calculations will be done for the three periods defined (2010-2016; 2017-2019; 2019-2022) and for the whole dataset.

Getting the data

To make this analysis we will depart from the files imported to Visone after adding the attributes (nuclear, ideology). They are part of the file “phases-visone-exports” and are displayed at the sheets “xxx-actors”.

Values for the whole dataset included in the graphs

In addition, we calculated the distribution per type of organization:

The calculations showed that 819 organizations are included in the graphs.

The point to highlight here is that not only federal agencies have been involved in the discussions, but a myriad of different organizations (directly or through different representatives). Media are the biggest group involved in the discussion, since it usually echoed and amplified the statements of other actors.

This graph does not measure participation in terms of number of statements, but only what kind of organizations have participated.

Participation per phases

calculate_metrics <- function(data) {
  # Calculation of the number of organizations
  orgs <- data %>%
    group_by(id) %>%
    summarise(organizaciones = n_distinct(name, na.rm = TRUE))
  n <- length(orgs$id)
  
  # Print the number of organizations
  cat("Number of organizations included in the graphs:", n, "\n\n")
  
  # Calculation of proportions by type
  result_by_type <- data %>%
    group_by(type) %>%
    summarise(Org_types = n_distinct(name, na.rm = TRUE))
  
  result_by_type <- result_by_type %>%
    arrange(desc(Org_types)) %>%
    mutate(type = factor(type, levels = type))
  
  # Create a chart
  pie_chart <- ggplot(result_by_type, aes(x = "", y = Org_types, fill = type)) +
    geom_bar(stat = "identity") +
    coord_polar("y", start = 0) +
    labs(title = "Proportion of Organization types in the whole dataset") +
    geom_text(aes(label = paste0(round(Org_types/sum(Org_types) * 100), "%")),
              position = position_stack(vjust = 0.5), size = 2) +
    theme_minimal() 
  
  print(pie_chart)
}

Phase 1

In phase 1, these were the organizations involved in the debate:

#calculations for phase 1
calculate_metrics(ph1)

## Number of organizations included in the graphs: 175

Phase 2

In phase 2, here are the results

#calculations for phase 2
calculate_metrics(ph2)

## Number of organizations included in the graphs: 358

Phase 3

#calculations for phase 3
calculate_metrics(ph3)

## Number of organizations included in the graphs: 474

DNA Analysis Addendum - Organization share

Rubens Yanes

2023-10-28