Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Help
Message posted by Lt. Dave Carlington on September 22, 2000 at 12:00 AM (ET)
I'am a police officer trying to compare monthly statistics(2-year period) for changes or reductions in false alarms based on a program we introduced. What is the basic method of comparison I could use for this data analysis? I have monthly stats 12 months pre and 12 months post reduction program efforts. Thanks a million! Lt DWC
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Help
Message posted by DL on September 22, 2000 at 12:00 AM (ET)
It looks as a paired means test, you just take month1 (before program) minus month1 (after program) and the resulting differences should be compared against null hipotesis = 0.
Re: Help
Message posted by jg on September 23, 2000 at 12:00 AM (ET)
Do time series analysis. Match month by month and allow for trend. Try to develop a monthly index for each month.
Re: Help
Message posted by jg on September 23, 2000 at 12:00 AM (ET)
That is you should have three effects - a one time change in average due to the new program, a mothly effect, and a possible trend effect.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.
