An Analysis of NYPD Arrest Data

Preregistration of analyses

Analysis #1

One analysis that we vow to make is to determine if there is a relationship between crime perpetrator characteristics and the level of offense they are charged with. Is there a potential for people of different races/sexes/ages being charged different levels of offense for a crime that falls under the same law code?

Null hypothesis race: The true proportion of black perpetrators arrested for felonies is the same white perpetrators

\[H_0: p_b = p_w\]

Alternative hypothesis race: The true proportion of black perpetrators arrested for felonies is not the the same as white perpetrators

\[H_A: p_b \neq p_w\]

Null hypothesis sex: The true proportion of male perpetrators arrested for felonies is the same as female perpetrators

\[H_0: p_m = p_f\]

Alternative hypothesis sex: The true proportion of male perpetrators arrested for felonies is not the same as female perpetrators

\[H_A: p_m \neq p_f\]

Null hypothesis offense: The true proportion of black perpetrators arrested for dangerous drugs who were convicted of felonies is the same as white perpetrators

\[H_0: p_b = p_w\]

Alternative hypothesis offense: The true proportion of black perpetrators arrested for dangerous drugs who were convicted of felonies is the same as white perpetrators

\[H_A: p_b \neq p_w\]

For all the above hypothesis we could use random sampling to create null distributions that identify whether the observed statistic is greater than or less than the p-value of 0.05.

Analysis #2

Another analysis we plan to make is to determine if there is a relationship between borough and crime perpetrator characteristics. Do certain boroughs have higher percentages of white people or women or old people committing crimes?

Null hypothesis: The true proportion of female perpetrators in Staten Island is the same as citywide

\[H_0: p_c = p_s\]

Alternative hypothesis: The true proportion of female perpetrators in Staten Island is not the same as citywide

\[H_A: p_c \neq p_s\]

Analysis #3

Are there trends between time of year and number of arrests made? Do these trends look different across different misdemeanors, races, boroughs, ages, or genders?

Null hypothesis: The average day of the month female perpetrators are arrested is the same as male perpetrators.

\[H_0: mu_f = mu_m\]

Alternative hypothesis: The average day of the month female perpetrators are arrested is not the same as male perpetrators.

\[H_A: mu_f \neq mu_m\]