Analysis of Parameter: Sub_Population_Target_z_6

Outline of the report

This automated report is divided into three sections, one for each failure condition.

Failure Condition 1 analyses the relation between the parameter: Sub_Population_Target_z_6, and the probability that the simulation resulted in no change, i.e. no reduction in the rate of FGM.

Failure Condition 2 analyses the relation between the parameter: Sub_Population_Target_z_6, and the probability that the simulation resulted in negative spillover, i.e. the number of people abandoning at the end of the simulation is less than the number initially targeted by the intervention.

Failure Condition 3 analyses the relation between the parameter: Sub_Population_Target_z_6, and the probability that the simulation only resulted in partial change, i.e. some people carried on practicing FGM at the end of the simulation.

The analysis used depends on whether the parameter is continuous or discrete

What to Expect: When the parameter is continuous

The first part of the analysis involves a technique called Monte Carlo Filtering (see Saltelli et al., who are cited in the thesis). In essence, this tests whether the cumulative distribution of the parameter varies between the situations where the failure scenario occurs and the situations where the failure scenario doesn’t occur. If it does, then this can be taken as an indication that the parameter has a marginal effect on the probability of the failure scenario. The test used to establish the difference is the KS-Test (with alpha set at 0.01)

The second part of the analysis involves a polynomial regression of the failure scenario on the parameter in question. This model is specified as

\[[Failure-Condition] = \beta_0 + \beta_1 [intervention-size] + \beta_2 [parameter] + \beta_3[parameter]^2 + \beta_4[parameter]^3 + \epsilon\] If one or more of the beta coefficients on the parameter is significant, this may indicate a relation between the parameter and the outcome.

The third part of the analysis is a visualization of the relation between the parameter value and the probability of the outcome, assuming an intervention of size 50%, based on the polynomial regression model.

What to Expect: When the parameter is discrete

When the parameter is discrete, the first part of the analysis is tabular. The simulation results are divided into groups based on intervention size, the groups are:

  1. All Interventions
  2. Intervention size \(\leq\) 25%%
  3. Intervention size > 25% and \(\leq\) 50%

For each group, the discrete parameter is cross-tabulated with the failure condition, to give a probability that the failure condition occurred under each parameter value.

Next, a regression model is reported. This model has the following specification:

\[[Failure-Condition] = \beta_0 + \beta_1 [intervention-size] + \beta_2 [parameter] + \beta_2[parameter]\cdot[intervention-size]\] It reports the main effects of the discrete parameter (controlling for intervention size), as well as potential interactions with the intervention size.

Next, a plot shows a linear visualization of the relationship between intervention size and the probability of the failure condition. A new gradient is fit for each discrete value of the parameter, allowing the viewer to visualize the main effects of the parameters (i.e. changing intercept of the line) and possible interactions with intervention size (i.e. changing gradient of the line)

Results for No Change (Failure Condition 1)

Cross Tab: Sub_Population_Target_z_6by Failure Condition 1 (All Simulations)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 1
1 low-h 1926 0.09
2 decision-makers 2030 0.10
3 by-family 2068 0.12
4 high-social-connectivity 1996 0.12
5 localised 1993 0.12
6 high-autonomy 1987 0.13
7 random 1986 0.13
8 high-authority 2014 0.14
9 high-family-connectivity 1990 0.14
10 anti-FGM-enforcers 1956 0.15
11 high-h 2026 0.16
12 pro-FGM-enforcers 2028 0.18
Cross Tab: Sub_Population_Target_z_6by Failure Condition 1 (Intervention Size <= 25%)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 1
1 low-h 476 0.12
2 decision-makers 524 0.15
3 by-family 513 0.16
4 pro-FGM-enforcers 1420 0.16
5 random 522 0.17
6 high-social-connectivity 495 0.19
7 anti-FGM-enforcers 1316 0.19
8 high-authority 515 0.20
9 high-autonomy 478 0.20
10 high-family-connectivity 499 0.20
11 high-h 519 0.21
12 localised 484 0.21
Cross Tab: Sub_Population_Target_z_6by Failure Condition 1 (Intervention Size > 25% and <= 50%)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 1
1 anti-FGM-enforcers 492 0.05
2 low-h 481 0.08
3 decision-makers 753 0.10
4 localised 516 0.11
5 high-autonomy 499 0.12
6 high-social-connectivity 487 0.13
7 random 461 0.13
8 by-family 506 0.13
9 high-authority 501 0.13
10 high-family-connectivity 510 0.15
11 high-h 478 0.18
12 pro-FGM-enforcers 485 0.20
Regression Model of Failure Condition 1 as a function of Sub
Dependent variable:
Failure Condition 1
initial-target-proportion -0.500***
(0.044)
Sub_Population_Target_z_6by-family -0.075***
(0.019)
Sub_Population_Target_z_6decision-makers -0.075***
(0.020)
Sub_Population_Target_z_6high-authority -0.045**
(0.019)
Sub_Population_Target_z_6high-autonomy -0.051***
(0.019)
Sub_Population_Target_z_6high-family-connectivity -0.030
(0.019)
Sub_Population_Target_z_6high-h -0.026
(0.019)
Sub_Population_Target_z_6high-social-connectivity -0.048**
(0.019)
Sub_Population_Target_z_6localised -0.038**
(0.019)
Sub_Population_Target_z_6low-h -0.137***
(0.019)
Sub_Population_Target_z_6pro-FGM-enforcers -0.097***
(0.016)
Sub_Population_Target_z_6random -0.068***
(0.019)
initial-target-proportion:Sub_Population_Target_z_6by-family 0.395***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6decision-makers 0.324***
(0.056)
initial-target-proportion:Sub_Population_Target_z_6high-authority 0.377***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6high-autonomy 0.374***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6high-family-connectivity 0.353***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6high-h 0.388***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6high-social-connectivity 0.347***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6localised 0.330***
(0.051)
initial-target-proportion:Sub_Population_Target_z_6low-h 0.460***
(0.052)
initial-target-proportion:Sub_Population_Target_z_6pro-FGM-enforcers 0.637***
(0.063)
initial-target-proportion:Sub_Population_Target_z_6random 0.409***
(0.051)
Constant 0.246***
(0.012)
Observations 24,000
R2 0.020
Adjusted R2 0.019
Residual Std. Error 0.336 (df = 23976)
F Statistic 21.078*** (df = 23; 23976)
Note: p<0.1; p<0.05; p<0.01

Results for Negative Spillover (Failure Condition 2)

Cross Tab: Sub_Population_Target_z_6by Failure Condition 2 (All Simulations)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 2
1 decision-makers 2030 0.30
2 anti-FGM-enforcers 1956 0.30
3 high-social-connectivity 1996 0.32
4 localised 1993 0.33
5 by-family 2068 0.34
6 random 1986 0.36
7 high-autonomy 1987 0.36
8 high-authority 2014 0.36
9 low-h 1926 0.37
10 high-family-connectivity 1990 0.37
11 pro-FGM-enforcers 2028 0.37
12 high-h 2026 0.37
Cross Tab: Sub_Population_Target_z_6by Failure Condition 2 (Intervention Size <= 25%)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 2
1 low-h 476 0.28
2 pro-FGM-enforcers 1420 0.32
3 decision-makers 524 0.32
4 high-social-connectivity 495 0.33
5 by-family 513 0.34
6 localised 484 0.34
7 anti-FGM-enforcers 1316 0.35
8 high-family-connectivity 499 0.36
9 random 522 0.37
10 high-h 519 0.37
11 high-autonomy 478 0.38
12 high-authority 515 0.41
Cross Tab: Sub_Population_Target_z_6by Failure Condition 2 (Intervention Size > 25% and <= 50%)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 2
1 anti-FGM-enforcers 492 0.22
2 decision-makers 753 0.27
3 localised 516 0.28
4 random 461 0.29
5 high-social-connectivity 487 0.29
6 low-h 481 0.31
7 high-autonomy 499 0.32
8 by-family 506 0.33
9 high-authority 501 0.34
10 high-family-connectivity 510 0.37
11 high-h 478 0.38
12 pro-FGM-enforcers 485 0.47
Regression Model of Failure Condition 2 as a function of Sub
Dependent variable:
Failure Condition 2
initial-target-proportion -0.518***
(0.062)
Sub_Population_Target_z_6by-family -0.095***
(0.027)
Sub_Population_Target_z_6decision-makers -0.100***
(0.028)
Sub_Population_Target_z_6high-authority -0.011
(0.027)
Sub_Population_Target_z_6high-autonomy -0.059**
(0.027)
Sub_Population_Target_z_6high-family-connectivity -0.041
(0.027)
Sub_Population_Target_z_6high-h -0.037
(0.027)
Sub_Population_Target_z_6high-social-connectivity -0.106***
(0.027)
Sub_Population_Target_z_6localised -0.096***
(0.027)
Sub_Population_Target_z_6low-h -0.178***
(0.027)
Sub_Population_Target_z_6pro-FGM-enforcers -0.132***
(0.023)
Sub_Population_Target_z_6random -0.080***
(0.027)
initial-target-proportion:Sub_Population_Target_z_6by-family 0.586***
(0.072)
initial-target-proportion:Sub_Population_Target_z_6decision-makers 0.502***
(0.078)
initial-target-proportion:Sub_Population_Target_z_6high-authority 0.452***
(0.072)
initial-target-proportion:Sub_Population_Target_z_6high-autonomy 0.548***
(0.072)
initial-target-proportion:Sub_Population_Target_z_6high-family-connectivity 0.524***
(0.073)
initial-target-proportion:Sub_Population_Target_z_6high-h 0.523***
(0.072)
initial-target-proportion:Sub_Population_Target_z_6high-social-connectivity 0.570***
(0.072)
initial-target-proportion:Sub_Population_Target_z_6localised 0.570***
(0.072)
initial-target-proportion:Sub_Population_Target_z_6low-h 0.794***
(0.073)
initial-target-proportion:Sub_Population_Target_z_6pro-FGM-enforcers 1.006***
(0.089)
initial-target-proportion:Sub_Population_Target_z_6random 0.590***
(0.072)
Constant 0.404***
(0.017)
Observations 24,000
R2 0.011
Adjusted R2 0.010
Residual Std. Error 0.473 (df = 23976)
F Statistic 11.477*** (df = 23; 23976)
Note: p<0.1; p<0.05; p<0.01

Results for Partial Change (Failure Condition 3)

Cross Tab: Sub_Population_Target_z_6by Failure Condition 3 (All Simulations)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 3
1 high-h 2026 0.67
2 high-authority 2014 0.79
3 high-autonomy 1987 0.80
4 random 1986 0.90
5 decision-makers 2030 0.90
6 by-family 2068 0.90
7 pro-FGM-enforcers 2028 0.91
8 localised 1993 0.91
9 high-family-connectivity 1990 0.91
10 high-social-connectivity 1996 0.93
11 low-h 1926 0.95
12 anti-FGM-enforcers 1956 0.96
Cross Tab: Sub_Population_Target_z_6by Failure Condition 3 (Intervention Size <= 25%)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 3
1 high-h 519 0.87
2 pro-FGM-enforcers 1420 0.91
3 high-authority 515 0.93
4 high-autonomy 478 0.93
5 decision-makers 524 0.96
6 high-family-connectivity 499 0.96
7 random 522 0.97
8 low-h 476 0.97
9 by-family 513 0.97
10 high-social-connectivity 495 0.97
11 localised 484 0.97
12 anti-FGM-enforcers 1316 0.97
Cross Tab: Sub_Population_Target_z_6by Failure Condition 3 (Intervention Size > 25% and <= 50%)
Value of Sub_Population_Target_z_6 N Probability of Failure Condition 3
1 high-h 478 0.74
2 high-authority 501 0.83
3 high-autonomy 499 0.85
4 pro-FGM-enforcers 485 0.90
5 decision-makers 753 0.90
6 random 461 0.92
7 high-family-connectivity 510 0.93
8 by-family 506 0.94
9 anti-FGM-enforcers 492 0.94
10 localised 516 0.94
11 low-h 481 0.95
12 high-social-connectivity 487 0.96
Regression Model of Failure Condition 3 as a function of Sub
Dependent variable:
Failure Condition 3
initial-target-proportion -0.195***
(0.041)
Sub_Population_Target_z_6by-family 0.004
(0.018)
Sub_Population_Target_z_6decision-makers -0.003
(0.019)
Sub_Population_Target_z_6high-authority -0.013
(0.018)
Sub_Population_Target_z_6high-autonomy -0.014
(0.018)
Sub_Population_Target_z_6high-family-connectivity -0.007
(0.018)
Sub_Population_Target_z_6high-h -0.068***
(0.018)
Sub_Population_Target_z_6high-social-connectivity 0.012
(0.018)
Sub_Population_Target_z_6localised 0.018
(0.018)
Sub_Population_Target_z_6low-h -0.019
(0.018)
Sub_Population_Target_z_6pro-FGM-enforcers -0.086***
(0.015)
Sub_Population_Target_z_6random -0.001
(0.018)
initial-target-proportion:Sub_Population_Target_z_6by-family 0.001
(0.048)
initial-target-proportion:Sub_Population_Target_z_6decision-makers -0.034
(0.052)
initial-target-proportion:Sub_Population_Target_z_6high-authority -0.197***
(0.048)
initial-target-proportion:Sub_Population_Target_z_6high-autonomy -0.167***
(0.048)
initial-target-proportion:Sub_Population_Target_z_6high-family-connectivity 0.040
(0.048)
initial-target-proportion:Sub_Population_Target_z_6high-h -0.319***
(0.047)
initial-target-proportion:Sub_Population_Target_z_6high-social-connectivity 0.036
(0.048)
initial-target-proportion:Sub_Population_Target_z_6localised -0.009
(0.048)
initial-target-proportion:Sub_Population_Target_z_6low-h 0.138***
(0.048)
initial-target-proportion:Sub_Population_Target_z_6pro-FGM-enforcers 0.185***
(0.059)
initial-target-proportion:Sub_Population_Target_z_6random -0.004
(0.048)
Constant 0.995***
(0.011)
Observations 24,000
R2 0.107
Adjusted R2 0.106
Residual Std. Error 0.313 (df = 23976)
F Statistic 125.261*** (df = 23; 23976)
Note: p<0.1; p<0.05; p<0.01