{"query_echo":{"q":"","filters":{"analysis_type":["multivariate regression"]},"warnings":[{"code":"facet_value_not_in_vocabulary","message":"Value 'multivariate regression' for facet 'analysis_type' is not in the controlled vocabulary."}]},"total":{"value":1,"relation":"eq"},"page":1,"page_size":20,"self":"https://climateinsightsdirectory.org/api/v1/search?analysis_type=multivariate+regression","stability":"beta","ai_marking":{"digital_source_type":{"uri":"http://cv.iptc.org/newscodes/digitalsourcetype/trainedAlgorithmicMedia","name":"Created using Generative AI"},"notice":"Summary and key insights are generated using artificial intelligence."},"results":[{"id":"prio-001714","url":"https://climateinsightsdirectory.org/document/prio-001714","title":"Running on Faith?","publisher":{"name":"Peace Research Institute Oslo","organization":"peace-research-institute-oslo"},"document_type":"research paper","language":"en","date_published":"2008-11","source_url":"https://cdn.cloud.prio.org/files/084c7220-6497-4243-a8e6-afbcb0e6d1a6/Running on Faith - A Quantitative Analysis.pdf?inline=true","generator":"Climate Insights Directory","generated_fields":["summary","key_insights","geography_discussed","analysis_type"],"summary":{"text":"The study finds that conflicts with religious cleavages are significantly more intense than non-religious ones. The effect on duration is ambiguous: religious conflicts are more likely to terminate in their early stages but are harder to end once they have lasted at least two and a half years.","ai_generated":true,"unit":"prio-001714#summary","date_modified":"2026-10-23"},"key_insights":[{"text":"Conflicts characterized by identity-based religious cleavages are significantly more violent than those without such cleavages. Multivariate analysis indicates that religious conflicts are predicted to have a 39.1% to 48.7% increase in annual battle-related deaths compared to non-religious conflicts, ceteris paribus.","ai_generated":true,"unit":"prio-001714#insight-1","date_modified":"2026-10-23"},{"text":"The impact of religious cleavages on conflict duration is time-dependent. In the early phases, religious conflicts are more likely to be terminated; however, after approximately two and a half years, they become less likely to end than non-religious conflicts.","ai_generated":true,"unit":"prio-001714#insight-2","date_modified":"2026-10-23"},{"text":"Religious polarization is a significant predictor of both intensity and duration. Higher levels of religious polarization are associated with increased conflict intensity and a higher likelihood of prolonged conflict.","ai_generated":true,"unit":"prio-001714#insight-3","date_modified":"2026-10-23"},{"text":"Religious fractionalization is associated with higher conflict intensity, contradicting the hypothesis that it would reduce intensity. It does not have a statistically significant effect on conflict duration.","ai_generated":true,"unit":"prio-001714#insight-4","date_modified":"2026-10-23"},{"text":"Religious discrimination tends to prolong religious conflicts. In religious conflicts, higher levels of discrimination are associated with a reduced risk of termination.","ai_generated":true,"unit":"prio-001714#insight-5","date_modified":"2026-10-23"},{"text":"The effect of religious legitimacy (defined as the presence of a state religion) on intensity varies by conflict type. In non-religious conflicts, high religious legitimacy predicts higher intensity, whereas in religious conflicts, it is associated with lower intensity levels.","ai_generated":true,"unit":"prio-001714#insight-6","date_modified":"2026-10-23"},{"text":"Religious legitimacy also has a time-varying effect on duration. It is predicted to increase the risk of termination in the first 234 days of a religious conflict, but it prolongs the conflict after that period.","ai_generated":true,"unit":"prio-001714#insight-7","date_modified":"2026-10-23"},{"text":"Regime type has a non-linear, non-monotonic effect on conflict intensity. The highest intensity levels are expected in slightly authoritarian regimes (Polity score of -2.75), while fully democratic states are predicted to have 52.5% less intense conflicts than fully authoritarian states.","ai_generated":true,"unit":"prio-001714#insight-8","date_modified":"2026-10-23"},{"text":"Conflict history is a strong predictor of intensity, as indicated by a positive and highly significant coefficient for the lagged battle-deaths variable.","ai_generated":true,"unit":"prio-001714#insight-9","date_modified":"2026-10-23"},{"text":"The number of simultaneous conflicts in a country is a strong predictor of overall intensity levels, as multiple conflicts lead to more battles and deaths.","ai_generated":true,"unit":"prio-001714#insight-10","date_modified":"2026-10-23"}],"climate_tags":[],"geography_discussed":["Democratic Republic of the Congo","Sri Lanka"],"frame_perspective":[],"analysis_type":["quantitative analysis","multivariate regression","survival analysis"]}]}