Page 238 - RUICHSS 2023 Proceeding

University of Ruhuna ISSN: 2706-0063
Matara, Sri Lanka

country i. is the intercept of the regression equation. is the coefficient
0
1
for the linear variable of URB1 indicating how much the dependent variable

changes for one-unit change in URB1. is the coefficient for the squared
2
term of URB1, indicating the impact of the quadratic (squared) relationship

between URB1 and the dependent variable.

At the initial phase of the study, scatter plots were constructed using the data
that were collected for the two variables of urban population as a percentage

of the total population and number of people with a mental disorder as a

percentage of the total population for the period of 1990-2019. From the
constructed scatter plot lines, the best-fit line needed to be identified, which

was suggested by the polynomial regression model. This was a clear choice
because the estimation taken from the linear relationship when compared to

the estimation taken from the quadratic polynomial regression the R squared

value was comparatively higher for the polynomial model. Thus it was utilised
to illustrate all of the countries.

7. Results and Discussion

Table 1 and 2 provides descriptive statistics for the two variables (urbanisation
and mental disorders) in SAARC countries. Table 2 shows the number of

observations (Obs), mean, standard deviation (SD), minimum (min), and
maximum (max) values of each variable in each SAARC country. From 1990

to 2019, there were 232 total observations with 30 observations corresponding
to SAARC countries.
179