And none of what they did addresses the root cause of the problem. that you can’t take two campaigns in the same ‘region’ and separate out the impact.

Since they didnt provide the actual data they analyzed nothing they said has any value or merit.

Edit: Let me make this simple for you

To estimate the effects of each independent variable on the likelihood of cam- paign success, we employed multinomial logistic regression (MLR), which compares the probabilities that different independent variables will result in each respective outcome: success, limited success, or failure.

Even the math they used to generate their results is incorrect. two campaigns (one non-violent, one violent) in the same political sphere are not independent variables.

Here, I opened the book (research paper) for you. and look! violence was involved in the successful outcomes!

especially in light of the enduring violent insurgencies occurring in some of the same countries—begs systematic investigation.

In many cases, both nonviolent and violent campaigns exist simultaneously among different competing groups.

you can’t isolate the non-violent campaigns from the violent campaigns in such an analysis, they’re part of the same environment.

I’ll also note that the research paper doesnt clearly identify how they generated their numbers. the underlying dataset they used is enitrely absent. so we can’t make any inferences on how well they actually isolated and calculated their values.

in short: doesnt show its work.