4 HOMEWORKPOR RESS ANYLS

profilexmamprosf

regression analyses response

DUE SATURDAY 2/7/15 by 6:00 pm New Yok Time

Responds to these two Posts in a paragrah or more by providing feedback on the regression analyses presented in their posts. Are these regression analyses appropriate? Why or why not? If not, offer suggestions and/or ideas about other analytical tools you think are more appropriate. These do not have to ellobrate but basic level engagement

 

 BOBThe best method of evaluation for the project chosen would be the binary logistic regression method.  The research needed for the study to make sure the New Harbor Port Authority is not racially profiling would need to be dichotomous.  The approach needs to be concise and to the point with no lingering thoughts left to build upon. (Garson, 2009)  The answer needed would need to be yes or no.  The binary logistic regression method goes through the numbers and puts the information on prospective of a positive or a negative. (Johnson, 2014)  With the port authority they need to make sure either they are racially profiling or not racially profiling.  All of the information can be calculated to determine to a number to find out the answer.  Unfortunately if the port authority is racially profiling the numbers will not lie nor will they lie if the port authority is doing the right thing. (Johnson, 2014)

            The binary regression method is a definitive and final answer based on research numbers provided.  Take the numbers and apply them to see if the answer is a positive or a negative.  The binary numbers or the 0’s or 1’s of the research will show the answer if it is a good answer or a not so good answer.  The method used for this study is to define racial profiling.  The answer needs to be a clear yes or no. (Garson, 2009)  There can be no variables such as times of day or das of the week.  The answer needs to be definitive with a numbers stating yes or no.  Without the definitive there will still be a lot of questions to answer.  If there is any opportunity to leave anything in the open it will be exploited.  This is why the binary regression method would be the best for this study. (Johnson, 2014)

 

Resources

Garson, G. D. (2009, April 4). Logistic regression. Retrieved from http://faculty.chass.ncsu.edu/garson/PA765/logistic.htm

Johnson, G. (2014). Research methods for public administrators (3rd Ed.). Armonk, NY: M. E. Sharpe.

Laureate Education (Producer). (2013c). Virtual community. [Multimedia file]. Baltimore, MD: Author

 

2. Mr. Tanner

Binary logistics regression would be the most accurate form of evaluation for my research project. Binary logistic regression is “a form of regression which is used when the dependent is a dichotomy and the independents are of any type”(Garson, 2009, para#1).  When trying to decide rather there is a decrease in the prison system and recidivism there are several variables that contribute. The Binary logistics evaluates division and contrast between two things that are represented as entirely different. Binary logistics works with multiple regression statistics. According to Johnson (2014), “this analytical technique measures several independent variables simultaneously to explain the changes in the dependent variable” (p.219). There are several variables such as jobs, gender, race, support systems, environment, etc. that affect the outcome of rather recidivism decreases and rather prison crowding decrease.

 

There are limitations to binary logistics. Correlation between some independent variables will “throw off the regression model, which works best when the independent variables are not correlated to each other” (Johnson, 2014, p.226). Therefore I would have to be careful when utilizing arrest records against crime rates. They are correlated with one another which can skew the data regarding reduction in crime rates.

 

References:

Johnson, G. (2014). Research methods for public administrators (3rd ed.). Armonk, NY: M. E. Sharpe.

Garson, G. D. (2009, April 4). Logistic regression. Retrieved from  http://faculty.chass.ncsu.edu/garson/PA765/logistic.htm

 

    • 9 years ago
    • 20
    Answer(0)
    Bids(0)