406_Conflict_db_analyses3
Analyses after meeting 23rd March 2023, taking into account comments of 8th March
- Proximate drivers
- Most common drivers
- How do drivers vary spatially?
- Which drivers co-occur?
- How are drivers and conflict types related?
- Drivers -> conflict number - with an eye towards potential lumping
(Devlin et al. -> drivers of conflict)
- Distal drivers
- Most common drivers
- How do drivers vary spatially?
- Which drivers co-occur?
- How are drivers and conflict types related?
- Approach:
Focus (so far in this doc): proximate drivers to conflict types, and how
Alternative to the clustering approach:
Calculate number of occurences for combinations of proximate drivers and conflict types
Make a “Sankey” plot to show connections drivers -> conflict types
Doesn’t work as well for actors - there are many combinations (even after lumping different ‘government’ and ‘ngo’ groups together). Perhaps rather use these as predictors for the different confict types directly (not as combinations)
Will try same approach for violence/illegality and for resolution
Data
Main data (ce)
[1] 1089 111Checking data
[1] 235[1] 49$`Article title (article_titl)`
[1] "Coast Guard Arrests 'Notorious' Toothfish Poacher, Minister Ziankahn Confirms"
$`Author (author) \r\nIf no author, leave blank.`
[1] "-999"
$`Source (source)`
[1] "AllAfrica"
$`Year (yr)`
[1] 2018
$`Month (month)`
[1] 3
$`Day\r\n(day)`
[1] 22
$`Language (lang)`
[1] "E"
$`ISO3 country code (iso3_code)`
[1] "LBR"Extract short names
Make ID column
- conf_event_id has some duplicates
- make a new version
==== conf_event_id ==== The following numbers should be 0 if there are no duplicates: [1] 0.02159624[1] 23==== conf_event_id2 ==== The following numbers should be 0 if there are no duplicates: [1] 0[1] 0Get relatively binary variables
Plot tables
Pick columns
- Not including ‘Resolution’ variables
- Data ‘ce2’
Make violence binary
Remove rows with piracy
Type = piracy: 137 rows will be removed Actor = pirates: 4 rows will be removed Function for summarizing columns
Proximate drivers
Pick variables
[1] "conf_event_id2" "benef_dist_driv" "acut_food_short"
[4] "nat_dis" "trad_v_new" "bord_disp_prox"
[7] "inad_gov_prox" "lack_reg_prox" "pol_insta_prox"
[10] "res_cond_hum_prox" "res_cond_nonhum_prox" "ill_res_use_prox"
[13] "iuu_fish_prox" "piracy_prox" "plan_dev" Summarize combinations
- Summarise the number of rows with a given variable or combination of variables (drivers in this case)
- I.e., the number of cases where the value of that variable equals 1
- I.e., the number of cases where the value of that variable equals 1
- Sort from the most common to least common combination of variables
- Make a new text variable which summarizes the names of the variables
- Make new data set with the new text variable and number of rows, add cumulative percentage
Plot
[1] "benef_dist_driv: Distribution of benefits (benef_dist_driv)"
[2] "bord_disp_prox: New or recent border dispute with neighbors (bord_disp_prox)"
[3] "inad_gov_prox: Recent change resulting in inadequate governance or enforcement due to a lack of capacity or resources or to a dispute (inad_gov_prox)"
[4] "lack_reg_prox: Recent change resulting in inadequate, lacking, or conflicting regulation (lack_reg_prox)"
[5] "res_cond_hum_prox: Recent change resulting in poor resource condition due to human activity (e.g. recent bout of overfishing or pollution) (res_cond_hum_prox)"
[6] "res_cond_nonhum_prox: Recent change resulting in poor resource condition due to non-human factors, excluding natural disasters (res_cond_nonhum_prox)"
[7] "ill_res_use_prox: Specific instance(s) of illegal use of a resource (e.g. illegal mangrove harvesting), excluding IUU fishing (ill_res_use_prox)"
[8] "iuu_fish_prox: Specific instance(s) of IUU fishing (iuu_fish_prox)"
[9] "plan_dev: Planned development (plan_dev)" Summarize columns but not rows
Distal drivers
Pick variables
[1] "conf_event_id2" "poverty" "soc_eco_ineq"
[4] "food_insec_dist" "clim_chang" "bord_disp_dist"
[7] "ethnic_div_dist" "cons_dem" "pop_inc"
[10] "inad_gov_dist" "pol_insta_dist" "poor_res_hum_dist"
[13] "poor_res_no_hum_dist" "marit_crime_dist" "ill_res_use_dist"
[16] "iuu_fish_dist" Summarize combinations
- Summarise the number of rows with a given variable or combination of variables (drivers in this case)
- I.e., the number of cases where the value of that variable equals 1
- I.e., the number of cases where the value of that variable equals 1
- Sort from the most common to least common combination of variables
- Make a new text variable which summarizes the names of the variables
- Make new data set with the new text variable and number of rows, add cumulative percentage
Plot
[1] "poverty: Poverty (poverty)"
[2] "clim_chang: Climate change (clim_chang)"
[3] "bord_disp_dist: Ongoing border dispute with neighbors (bord_disp_dist)"
[4] "inad_gov_dist: Ongoing inadequate governance or enforcement (inad_gov_dist)"
[5] "pol_insta_dist: Ongoing political instability (pol_insta_dist)"
[6] "poor_res_hum_dist: Poor resource condition due to human activity, ongoing (poor_res_hum_dist)"
[7] "poor_res_no_hum_dist: Poor resource condition due to non-human factors, ongoing (poor_res_no_hum_dist)"
[8] "marit_crime_dist: Ongoing illegal activity or maritime crime, excluding piracy, illegal use of resources, or IUU fishing (marit_crime_dist)"
[9] "ill_res_use_dist: Systemic illegal use of a resource (e.g. illegal mangrove harvesting), excluding IUU fishing (ill_res_use_dist)"
[10] "iuu_fish_dist: Systematically occurring IUU fishing (iuu_fish_dist)" Conflict types
Pick variables
[1] "conf_event_id2" "access_space" "use_space" "access_res"
[5] "use_res" "benef_dist_type" "piracy" Summarize combinations
- Summarise the number of rows with a given variable or combination of variables (drivers in this case)
- I.e., the number of cases where the value of that variable equals 1
- I.e., the number of cases where the value of that variable equals 1
- Sort from the most common to least common combination of variables
- Make a new text variable which summarizes the names of the variables
- Make new data set with the new text variable and number of rows, add cumulative percentage
Most common co-occurences
| column_comb | n | percent | cumul_percent |
|---|---|---|---|
| access_space_plus | 238 | 25.1 | 25.1 |
| access_space_only | 212 | 22.4 | 47.5 |
| access_res_only | 182 | 19.2 | 66.7 |
| use_space_only | 132 | 13.9 | 80.6 |
| use_res_only | 75 | 7.9 | 88.5 |
| access_res_plus | 53 | 5.6 | 94.1 |
| other | 44 | 4.6 | 98.7 |
| use_space_plus | 12 | 1.3 | 100.0 |
Plot
character(0)Summarize columns but not rows
Sankey diagram, prox + type
NOTE: This plot doesn’t plot well in pdf
Including “None” group of prox
Static plot
Excluding ‘none’ group
addNA(from)
benef_dist_driv benef_dist_driv + lack_reg_prox
91 28
iuu_fish_prox lack_reg_prox
263 52
None other_prox
272 162
plan_dev res_cond_hum_prox
28 52
<NA>
0 Actors - combinations
Pick variables
[1] "conf_event_id2" "local_ngo" "natl_ngo" "intl_ngo"
[5] "loc_govt" "intl_govt" "subs_fisher" "ind_fishers"
[9] "fish_trad" "subs_living_res" "ind_living_res" "ind_non_liv"
[13] "develop" "comm_org" "worker_union" "pirates"
[17] "state_enf" "intl_multi_bod" Summarize combinations
- Summarise the number of rows with a given variable or combination of variables (drivers in this case)
- I.e., the number of cases where the value of that variable equals 1
- I.e., the number of cases where the value of that variable equals 1
- Sort from the most common to least common combination of variables
- Make a new text variable which summarizes the names of the variables
- Make new data set with the new text variable and number of rows, add cumulative percentage
Check ‘intl_govt + fishers’
- Percentages
local_ngo natl_ngo intl_ngo loc_govt intl_govt
0.0 0.0 0.0 0.0 100.0
subs_fisher ind_fishers fish_trad subs_living_res ind_living_res
70.7 60.7 8.3 2.7 0.3
ind_non_liv develop comm_org worker_union pirates
4.7 0.0 2.7 31.7 0.0
state_enf intl_multi_bod ngo column_comb
36.1 8.9 0.0 0.0 Most common co-occurences
| column_comb | n | percent |
|---|---|---|
| intl_govt + fishers | 338 | 35.7 |
| intl_govt | 131 | 13.8 |
| subs_fisher | 79 | 8.3 |
| intl_govt + fishers + ngo | 75 | 7.9 |
| intl_govt + ngo | 59 | 6.2 |
| ind_fishers | 52 | 5.5 |
| other | 42 | 4.4 |
| loc_govt + fishers | 41 | 4.3 |
| intl_govt + loc_govt + fishers | 30 | 3.2 |
| loc_govt | 21 | 2.2 |
| subs_fisher + ind_fishers | 18 | 1.9 |
| intl_govt + loc_govt | 16 | 1.7 |
| ind_fishers + ngo | 14 | 1.5 |
| intl_govt + loc_govt + ngo | 8 | 0.8 |
| subs_fisher + ngo | 7 | 0.7 |
| loc_govt + ngo | 6 | 0.6 |
| intl_govt + loc_govt + fishers + ngo | 5 | 0.5 |
| loc_govt + fishers + ngo | 3 | 0.3 |
| subs_fisher + ind_fishers + ngo | 3 | 0.3 |
Plot
[1] "local_ngo: Local NGO (local_ngo)"
[2] "natl_ngo: National NGO (natl_ngo)"
[3] "intl_ngo: International NGO (intl_ngo)"
[4] "loc_govt: Local or regional government (loc_govt)"
[5] "intl_govt: National government (intl_govt)"
[6] "subs_fisher: Small-scale, subsistence, or artisanal fishers (subs_fisher)"
[7] "ind_fishers: Industrial fishers (ind_fishers)"
[8] "fish_trad: Fish traders, wholesalers, and retailers (fish_trad)"
[9] "subs_living_res: Small-scale, subsistence, or recreational living resource users, except fishers (subs_living_res)"
[10] "ind_living_res: Industrial living resource users, except fishers (ind_living_res)"
[11] "ind_non_liv: Industrial non-living resource users (ind_non_liv)"
[12] "develop: Development actors (develop)"
[13] "comm_org: Community organization (comm_org)"
[14] "worker_union: Worker union or association (worker_union)"
[15] "pirates: Pirates (pirates)"
[16] "state_enf: State enforcement agents (state_enf)"
[17] "intl_multi_bod: International multilateral bodies (intl_multi_bod)" Summarize columns but not rows
Actors - simple statistics
| actor | n | percent |
|---|---|---|
| intl_govt | 662 | 21.7 |
| subs_fisher | 457 | 15.0 |
| ind_fishers | 385 | 12.6 |
| state_enf | 351 | 11.5 |
| ngo | 195 | 6.4 |
| worker_union | 157 | 5.1 |
| loc_govt | 130 | 4.3 |
| subs_living_res | 125 | 4.1 |
| intl_multi_bod | 108 | 3.5 |
| natl_ngo | 105 | 3.4 |
| ind_non_liv | 90 | 2.9 |
| intl_ngo | 87 | 2.8 |
| fish_trad | 68 | 2.2 |
| comm_org | 62 | 2.0 |
| local_ngo | 36 | 1.2 |
| develop | 25 | 0.8 |
| ind_living_res | 10 | 0.3 |
Actors rel. to each other
- For each row (actor), the percentage of cases of each actor
- First column = number of cases
Prox_driver N ...3 ...4 ...5 ...6 ...7 ...8 ...9 ...10 ...11 ...12 ...13
1 All 948 20.6 13.7 69.8 48.2 40.5 7.2 13.2 1.1 9.5 2.6 6.5
...14 ...15 ...16
1 16.6 37 11.4 Prox_driver N ngo loc_govt intl_govt subs_fisher ind_fishers
1 All 948 20.6 13.7 69.8 48.2 40.5
2 ngo 195 100 11.3 75.4 34.4 36.9
3 loc_govt 130 16.9 100 45.4 47.7 29.2
4 intl_govt 662 22.2 8.9 100 48.2 41.5
5 subs_fisher 457 14.7 13.6 69.8 100 38.5
6 ind_fishers 384 18.8 9.9 71.6 45.8 100
7 fish_trad 68 10.3 16.2 70.6 72.1 23.5
8 subs_living_res 125 26.4 25.6 64 24 2.4
9 ind_living_res 10 20 30 50 20 20
10 ind_non_liv 90 26.7 13.3 76.7 37.8 13.3
11 develop 25 60 28 68 12 0
12 comm_org 62 40.3 32.3 75.8 37.1 8.1
13 worker_union 157 8.3 9.6 89.8 82.8 49.7
14 state_enf 351 12.3 11.7 52.7 49 47
15 intl_multi_bod 108 19.4 14.8 82.4 32.4 44.4
fish_trad subs_living_res ind_living_res ind_non_liv develop comm_org
1 7.2 13.2 1.1 9.5 2.6 6.5
2 3.6 16.9 1 12.3 7.7 12.8
3 8.5 24.6 2.3 9.2 5.4 15.4
4 7.3 12.1 0.8 10.4 2.6 7.1
5 10.7 6.6 0.4 7.4 0.7 5
6 4.2 0.8 0.5 3.1 0 1.3
7 100 5.9 2.9 2.9 0 2.9
8 3.2 100 4.8 16.8 7.2 19.2
9 20 60 100 10 10 0
10 2.2 23.3 1.1 100 1.1 20
11 0 36 4 4 100 44
12 3.2 38.7 0 29 17.7 100
13 15.9 5.1 1.3 3.8 0.6 3.2
14 5.1 8.8 1.4 6 1.1 2.3
15 6.5 2.8 0 9.3 0.9 5.6
worker_union state_enf intl_multi_bod
1 16.6 37 11.4
2 6.7 22.1 10.8
3 11.5 31.5 12.3
4 21.3 27.9 13.4
5 28.4 37.6 7.7
6 20.3 43 12.5
7 36.8 26.5 10.3
8 6.4 24.8 2.4
9 20 50 0
10 6.7 23.3 11.1
11 4 16 4
12 8.1 12.9 9.7
13 100 17.8 7
14 8 100 6.6
15 10.2 21.3 100Actors rel. to prox drivers
- For each row (proximate driver), the percentage of cases of each actor
- First column = number of cases, last column = overall p-value for chi-sq. test of that row
Prox_driver N ngo loc_govt intl_govt subs_fisher ind_fishers
1 All 948 20.6 13.7 69.8 48.2 40.5
2 benef_dist_driv 160 16.9 11.2 81.9 66.2** 48.8
3 acut_food_short 10 20 30 90 90 30
4 nat_dis 1 0 100* 0 0 0
5 trad_v_new 6 16.7 50 83.3 100 0*
6 bord_disp_prox 20 0* 25 80 35 15
7 inad_gov_prox 35 11.4 22.9 60 74.3* 37.1
8 lack_reg_prox 100 20 12 91** 67** 37
9 pol_insta_prox 1 0 0 100 0 0
10 res_cond_hum_prox 85 36.5** 25.9* 67.1 38.8* 20***
11 res_cond_nonhum_prox 7 0 14.3 71.4 71.4 28.6
12 ill_res_use_prox 26 15.4 30.8* 42.3 19.2* 7.7**
13 iuu_fish_prox 301 11.6*** 10.6 52.5*** 46.2 68.1***
14 plan_dev 64 35.9* 18.8 87.5 35.9* 17.2***
fish_trad subs_living_res ind_living_res ind_non_liv develop comm_org
1 7.2 13.2 1.1 9.5 2.6 6.5
2 13.1** 6.2** 0.6 15* 3.1 8.8
3 40*** 0 0 0 0 10
4 0 100* 0 0 0 0
5 16.7 16.7 0 0 0 33.3*
6 0 0 0 10 0 15
7 25.7*** 17.1 0 2.9 0 2.9
8 12 6* 2 1** 0 2
9 0 0 0 0 0 0
10 4.7 35.3*** 3.5* 40*** 14.1*** 23.5***
11 14.3 14.3 0 14.3 14.3* 0
12 19.2* 57.7*** 11.5*** 7.7 7.7 15.4
13 4* 2*** 0* 0.7*** 0.3** 1.3***
14 0* 17.2 0 45.3*** 25*** 26.6***
worker_union state_enf intl_multi_bod p_value
1 16.6 37 11.4 <NA>
2 28.7*** 14.4*** 15.6 <0.0001
3 40 10 10 0.045
4 0 100 0 0.34
5 16.7 50 33.3 0.35
6 10 20 60*** <0.0001
7 25.7 31.4 8.6 0.0038
8 41*** 7*** 9 <0.0001
9 0 0 100*** 0.43
10 5.9** 23.5** 2.4** <0.0001
11 14.3 14.3 0 0.71
12 7.7 50 3.8 <0.0001
13 7.6*** 69.8*** 8.6 <0.0001
14 10.9 20.3** 4.7* <0.0001Actors rel. to conflict types
- For each row (conflict type), the percentage of cases of each actor
- First column = number of cases, last column = overall p-value for chi-sq. test of that row
Prox_driver N ngo loc_govt intl_govt subs_fisher ind_fishers
1 All 946 20.6 13.6 69.9 48.2 40.5
2 access_space 450 15.6*** 13.1 71.8 54.4* 50.4***
3 use_space 195 33.8*** 19 77.4 29.7*** 16.4***
4 access_res 416 20.7 13.5 65.9 55.8** 53.1***
5 use_res 229 21.4 16.2 74.2 51.5 62***
6 benef_dist_type 129 16.3* 12.4 83.7 75.2*** 50.4
fish_trad subs_living_res ind_living_res ind_non_liv develop comm_org
1 7.2 13.2 1.1 9.5 2.6 6.6
2 3.3*** 8*** 0.7 5.6*** 3.1 4.7*
3 3.6* 36.9*** 1.5 33.3*** 10.8*** 23.1***
4 10.3** 7.5*** 0.5 1*** 0.2*** 2.6***
5 9.2 8.3** 1.7 1.7*** 0.9* 4.4
6 20.2*** 7** 1.6 10.1 3.1 4.7
worker_union state_enf intl_multi_bod p_value
1 16.6 36.9 11.4 <NA>
2 18.2 46*** 12.2 <0.0001
3 11.3* 14.4*** 5.1** <0.0001
4 19.2 38.7 12 <0.0001
5 18.8 28.8** 14.8 <0.0001
6 38.8*** 9.3*** 10.9 <0.0001
Pearson's Chi-squared test
data: M
X-squared = 30.07, df = 2, p-value = 2.954e-07 party
gender Democrat Independent Republican
F 762 327 468
M 484 239 477 party
gender Democrat Independent Republican
F 703.6714 319.6453 533.6834
M 542.3286 246.3547 411.3166 party
gender Democrat Independent Republican
F 2.1988558 0.4113702 -2.8432397
M -2.5046695 -0.4685829 3.2386734 party
gender Democrat Independent Republican
F 4.5020535 0.6994517 -5.3159455
M -4.5020535 -0.6994517 5.3159455Actors rel. to conflict types (one table per driver)
Actors rel. to conflict types involving iuu_fish_prox
- For proximate driver = ‘iuu_fish_prox’ only
- For each row (conflict type), the percentage of cases of each actor
- First column = number of cases, last column = overall p-value for chi-sq. test of that row
Prox_driver N ngo loc_govt intl_govt subs_fisher ind_fishers fish_trad
1 All 300 11.7 10.7 52.3 46.3 68 4
2 access_space 191 12 9.4 53.9 48.2 71.2 1***
3 use_space 15 13.3 13.3 86.7 53.3 73.3 0
4 access_res 145 8.3 12.4 43.4* 48.3 70.3 4.8
5 use_res 102 16.7 16.7* 66.7* 43.1 78.4 5.9
6 benef_dist_type 15 20 26.7 66.7 100 73.3 20*
subs_living_res ind_living_res ind_non_liv develop comm_org worker_union
1 2 0 0.7 0.3 1.3 7.7
2 0.5* 0 0.5 0 1 8.9
3 6.7 0 0 6.7*** 6.7 13.3
4 3.4 0 0.7 0 2.1 8.3
5 1 0 0 0 2 7.8
6 20*** 0 0 0 0 40***
state_enf intl_multi_bod p_value
1 69.7 8.7 <NA>
2 75.4 7.9 NA
3 26.7** 20 NA
4 70.3 6.9 NA
5 45.1*** 14.7* NA
6 13.3*** 13.3 NAActors rel. to conflict types involving lack_reg_prox
- For proximate driver = ‘lack_reg_prox’ only
- For each row (conflict type), the percentage of cases of each actor
- First column = number of cases, last column = overall p-value for chi-sq. test of that row
Prox_driver N ngo loc_govt intl_govt subs_fisher ind_fishers
1 All 100 20 12 91 67 37
2 access_space 63 12.7* 7.9 98.4 76.2 42.9
3 use_space 5 20 60*** 60 20 0
4 access_res 53 24.5 15.1 86.8 66 49.1
5 use_res 20 25 10 100 75 50
6 benef_dist_type 23 17.4 13 82.6 69.6 34.8
fish_trad subs_living_res ind_living_res ind_non_liv develop comm_org
1 12 6 2 1 0 2
2 9.5 3.2 1.6 1.6 0 3.2
3 0 40*** 0 0 0 20***
4 13.2 7.5 0 0 0 3.8
5 10 0 0 0 0 0
6 26.1* 17.4* 4.3 0 0 0
worker_union state_enf intl_multi_bod p_value
1 41 7 9 <NA>
2 42.9 11.1 9.5 NA
3 0 0 20 NA
4 45.3 1.9* 11.3 NA
5 35 0 15 NA
6 60.9 0 0 NAActors rel. to conflict types involving res_cond_hum_prox
- For proximate driver = ‘res_cond_hum_prox’ only
- For each row (conflict type), the percentage of cases of each actor
- First column = number of cases, last column = overall p-value for chi-sq. test of that row
Prox_driver N ngo loc_govt intl_govt subs_fisher ind_fishers fish_trad
1 All 85 36.5 25.9 67.1 38.8 20 4.7
2 access_space 20 65 30 75 40 25 0
3 use_space 59 40.7 30.5 64.4 27.1** 11.9** 3.4
4 access_res 19 21.1 21.1 73.7 73.7** 31.6 15.8**
5 use_res 21 23.8 23.8 57.1 61.9* 47.6*** 0
6 benef_dist_type 7 71.4 57.1 85.7 42.9 28.6 14.3
subs_living_res ind_living_res ind_non_liv develop comm_org worker_union
1 35.3 3.5 40 14.1 23.5 5.9
2 25 5 30 35* 30 5
3 35.6 1.7 49.2* 20.3* 28.8 5.1
4 21.1 5.3 0** 0 10.5 21.1**
5 33.3 14.3** 9.5** 4.8 14.3 9.5
6 14.3 14.3 14.3 57.1* 42.9 0
state_enf intl_multi_bod p_value
1 23.5 2.4 <NA>
2 30 0 0.29
3 22 0* 0.00086
4 26.3 10.5** <0.0001
5 33.3 4.8 0.00024
6 14.3 0 0.22Illegal activity
Pick variables
[1] "conf_event_id2" "iuu_fish" "ill_spa_acc" "ill_spa_use"
[5] "ill_res_acc" "ill_res_use" "gear_destr" "infrast_destr"
[9] "ill_mar_crime" "ill_none" Summarize combinations
- Summarise the number of rows with a given variable or combination of variables (drivers in this case)
- I.e., the number of cases where the value of that variable equals 1
- I.e., the number of cases where the value of that variable equals 1
- Sort from the most common to least common combination of variables
- Make a new text variable which summarizes the names of the variables
- Make new data set with the new text variable and number of rows, add cumulative percentage
Most common co-occurences
| column_comb | n | percent |
|---|---|---|
| ill_none | 470 | 49.6 |
| iuu_fish | 333 | 35.1 |
| ill_spa_use | 45 | 4.7 |
| ill_res_use | 19 | 2.0 |
| ill_res_acc | 19 | 2.0 |
| ill_spa_acc | 14 | 1.5 |
| iuu_fish + gear_destr | 9 | 0.9 |
| iuu_fish + ill_mar_crime | 5 | 0.5 |
| None | 4 | 0.4 |
| ill_spa_acc + ill_spa_use | 4 | 0.4 |
| iuu_fish + infrast_destr | 3 | 0.3 |
| gear_destr | 2 | 0.2 |
| ill_res_acc + infrast_destr | 2 | 0.2 |
| iuu_fish + ill_none | 2 | 0.2 |
| iuu_fish + ill_res_use | 2 | 0.2 |
| iuu_fish + ill_spa_use | 2 | 0.2 |
| iuu_fish + ill_spa_acc | 2 | 0.2 |
| ill_mar_crime | 1 | 0.1 |
| infrast_destr | 1 | 0.1 |
| ill_res_acc + ill_res_use | 1 | 0.1 |
Plot
[1] "iuu_fish: IUU fishing (iuu_fish)"
[2] "ill_spa_acc: Illegal access to a space, excluding IUU fishing (ill_spa_acc)"
[3] "ill_spa_use: Illegal use of a space, excluding IUU fishing (ill_spa_use)"
[4] "ill_res_acc: Illegal access to a resource, excluding IUU fishing (ill_res_acc)"
[5] "ill_res_use: Illegal use of a resource, excluding IUU fishing (ill_res_use)" Summarize columns but not rows
column_comb
gear_destr ill_mar_crime ill_none ill_res_acc ill_res_use
12 6 474 21 19
ill_spa_acc ill_spa_use infrast_destr iuu_fish
14 52 6 344 Sankey diagram driver - type - illegality
Including “None” group of prox
Excluding ‘None’ group
Resolution
Pick variables
N Y
911 37 ev_res
ev_stalemate N Y
N 618 293
Y 37 0, , ev_unres = N
ev_res
ev_stalemate N Y
N 2 292
Y 20 0
, , ev_unres = Y
ev_res
ev_stalemate N Y
N 616 1
Y 17 0 ev_unres_inab_cons
ev_unres -999 0 1
N 50 264 0
Y 304 8 322 ev_unres_inab_cons
ev_unres -999 0 1
N 50 264 0
Y 304 8 322 unres_inab_gov
ev_unres -999 0 1
N 50 264 0
Y 430 12 192 ev_unres_soc_facts
ev_unres -999 0 1
N 50 264 0
Y 583 15 36 [1] "conf_event_id2" "ev_res_conf_med" "ev_res_inc_gov"
[4] "ev_leg_court_res" "ransom" "ev_unres_inab_cons"
[7] "unres_inab_gov" "ev_unres_soc_facts" "stalemate_other"
[10] "resolved_other" "unresolved_other" Summarize combinations
- Summarise the number of rows with a given variable or combination of variables (drivers in this case)
- I.e., the number of cases where the value of that variable equals 1
- I.e., the number of cases where the value of that variable equals 1
- Sort from the most common to least common combination of variables
- Make a new text variable which summarizes the names of the variables
- Make new data set with the new text variable and number of rows, add cumulative percentage
Most common co-occurences
| column_comb | n | percent |
|---|---|---|
| ev_unres_inab_cons | 268 | 28.3 |
| ev_res_inc_gov | 194 | 20.5 |
| unresolved_other | 157 | 16.6 |
| unres_inab_gov | 127 | 13.4 |
| ev_unres_inab_cons + unres_inab_gov | 37 | 3.9 |
| ev_leg_court_res | 37 | 3.9 |
| stalemate_other | 28 | 3.0 |
| resolved_other | 21 | 2.2 |
| ev_res_inc_gov + ev_leg_court_res | 19 | 2.0 |
| unres_inab_gov + ev_unres_soc_facts | 15 | 1.6 |
| ev_unres_inab_cons + unres_inab_gov + ev_unres_soc_facts | 13 | 1.4 |
| ev_res_conf_med | 12 | 1.3 |
| ev_res_conf_med + ev_leg_court_res | 6 | 0.6 |
| ev_unres_soc_facts | 4 | 0.4 |
| ev_unres_inab_cons + ev_unres_soc_facts | 4 | 0.4 |
| ev_res_conf_med + ev_res_inc_gov | 4 | 0.4 |
| None | 2 | 0.2 |
Plot
[1] "ev_res_inc_gov: Increased governance or enforcement (ev_res_inc_gov)"
[2] "ev_leg_court_res: Legal or court resolution (ev_leg_court_res)"
[3] "ev_unres_inab_cons: Inability to reach consensus (ev_unres_inab_cons)"
[4] "unres_inab_gov: Inability to increase or change governance or enforcement (unres_inab_gov)"