I chose to use this website to collect my data because I am interested in analyzing data from several countries rather than data from several companies, primarily all located in the United States. Since I knew that I wanted to study the relationship of adolescent fertility rate and GAP per capita, I started collecting my sample by looking in the variable indexes for subjects that I thought could also be somehow related to those two. Also, since the adolescent fertility rate is variable that is difficult to track in a timely manner, I chose to use a year a little further back as my base year (2000).
I also chose 2000 as the base year for life expectancy at birth, total (yr), public spending on education, total (% of GAP), GAP per capita (constant 2000 US$), and artillery rate, total (births per women). When deciding which variables to compare, I also looked at the literacy rate, long-term unemployment, and contraceptive prevalence, but they could not be used because there was missing data. 4. My data has many variables from one point in time (2000) so it is cross-sectional data. My identifier variable, country, is in my first column in my data set, which means that it Just has the name of the country.
My dependent variable, adolescent fertility rate, is in the second column and is quantitative, cardinal (meaning the preference between the data's numerical values has a meaning, and they are not Just ranked or ordered), and continuous (meaning it takes on an uncountable set of possible values). I en rest AT ten columns are my Tour Independent variables . I knees variables are all quantitative, cardinal, and continuous. The third column contains my focal independent variable, GAP per capita measured in constant 2000 IIS$.
The fourth column contains life expectancy at birth, measured in years. Column five is public spending on education (% of GAP). The last column is the fertility rate (births ere women). 5. The adolescent fertility rate is my dependent variable, and GAP per capita, measured in constant 2000 IIS$ is my focal independent variable. I would assume, without examining the data that there will be a strong negative correlation between these variables, meaning that as GAP increases, the adolescent fertility rate will decrease.
This relationship makes since because countries with a higher GAP will be more prone to having contraceptives available, causing a lower adolescent fertility rate. On the other hand, however, countries with higher Gaps would have greater lath care available resulting in fewer miscarriages and so could possibly result in a higher adolescent fertility rate. The other independent variables I chose were life expectancy at birth, public spending on education, and total fertility rate.
Originally, I chose these variables along with others such as contraceptive prevalence and literacy rate because I thought they could be related to the adolescent fertility rate. The correlation between some variables is still uncertain, however. An example of this is when comparing public spending on education to the adolescent fertility rate o see if there is any correlation and although correlation does not imply causation, it would be noteworthy to see if there is a decrease in the adolescent fertility rate with an increase in public spending on education.
I thought that my other independent variables, life expectancy at birth, and the total fertility rate, would have a strong correlation to the adolescent fertility rate. It makes sense that if the life expectancy is lower, there will be a greater adolescent fertility rate, and if the overall fertility rate is ere high (or low) the adolescent fertility rate will be very high (or low). These reasoning may not hold however in some cases.
For example, countries such as Ethiopia with a very low life expectancy may have a lower adolescent fertility rate because is has a greater incidence of malnutrition, and so this could interfere with the fertility of females. My hypotheses outlined above are not certain, and testing is necessary to check for any possible correlations among the data whether they be positive, negative, or possibly there is no relationship among the data.