| dc.description.abstract | Suicide is a major public health problem
worldwide and moreover it is considered as a social
problem. In Sri Lanka, suicides have been a major social
and economic burden for the previous three decades.
Hence, developing a statistical analysis on suicides is a
fruitful study for the country. This case study aims at
identifying factors that mainly associated with suicides in
the selected area covered by twelve police stations under
Kelaniya Police Division. Secondary data were collected
from those police stations from 2013 to 2017. Addiction
to narcotic drugs, chronic disease, economic problems,
loss, love affairs, mental disorder and family disputes
were considered as explanatory variables. The
significantly associated explanatory variables were
identified using a univariate analysis and included in the
multivariate analysis to perform multinomial logistic
regression.Attained results implies that, civil status, age
and gender were identified as influenced factors for both
chronic disease and mental disorder. Economic problems
are influenced by age and gender. Civil status has an
impact on loss. Moreover, Education level, civil status,
gender and age were identified as having impacts on love
affairs.Due to encountered classification problems and
high number of coefficients of the factors were
insignificant in themodel, above dependent variables
were re-categorized based on the influence for suicide
which lead to fit a binary logistic regression model. Based
on the results, civil status, education level and occupation
and age were identified as influenced factors for suicides
from the final model with the accuracy of 72.6% correctly
classified | |
