Computer Assignment #4
1st T-Test: The first test compares the sex of the offender being 1= male and 2=female, and the number of substances ever used.
Hypothesis: In this test I hypothesize that males (1) will have a larger number of substances used compared to the females (2) population.
Logic: According to past criminal studies comparing males and females with drug use/abuse, males are more likely to have a larger problem with substance abuse and variety.
Data: According to the data set the p-value of the data; this was 0.004, is less then the alpha value of 0.05. We can conclude that since that the p-value is less than the alpha value there is correlation between sex and number of different substances used. According to the data males use a larger variety of substances than do females.
2nd T-Test: The second test compares the race of the offender: (1) being white (2) being black, to the number of different crimes committed in the past year.
Hypothesis: In this test I hypothesize that blacks will have a larger amount of crimes committed in the past year according to the data.
Logic: According to the data in previous studies, minorities are more likely to commit a larger amount of various crimes than compared to whites.
Data: According to the data set the p-value of the data comparing race and the number of crimes in a year, the p-value is greater than the alpha value of 0.05 which means that there is no correlation between the race of the offender and the number of crimes committed per year.
1st Correlation: In this correlation we are comparing scale of association with pro-social peers and number of different problem behaviors ever participated in.
Hypothesis: I hypothesize that there will be a correlation between number of different problem behaviors ever participated in and scale of association with pro-social peers.
Logic: The larger amount of pro-social peers should affect the amount of problem behaviors in a positive manner.
Data: According to the data there is a negative correlation between the tow variables. When pro-social peers scale increases, the problem behavior scale decreases. The correlation is weak because the p-value is 0. To have a strong correlation, you need a value of 0.5 or greater to be strong.