topic Re: Need robust method to calculate statistical difference in Statistical Procedures

Need robust method to calculate statistical difference

GARYV — Fri, 12 Jan 2018 15:47:53 GMT

Hello,

I am new to data analysis and need help with the following.

I have a survey data summary table where i have results from state and area of that state. I am interested in yes response

It looks like following for each question

Participants Yes C.I No C.I.

STATE 7450 88% 83%-93% 12% 10% -14%

Area 668 82 80%-84% 18% 16% -20%

How do I calculate statistical difference in above two and decide if its statistically significant?

I can do it by calculating yes and no response number from percentages given and do chi square test but it wont give a robust results.

Really appreciate input and help.

Re: Need robust method to calculate statistical difference

Reeza — Fri, 12 Jan 2018 15:57:29 GMT

Chi Square is robust...and your N is large enough to obtain reliable estimates.

If you want to explore non-parametric methods, look at PROC NPAR1WAY.

@GARYV wrote:

Hello,

I am new to data analysis and need help with the following.

I have a survey data summary table where i have results from state and area of that state. I am interested in yes response

It looks like following for each question

Participants Yes C.I No C.I.

STATE 7450 88% 83%-93% 12% 10% -14%

Area 668 82 80%-84% 18% 16% -20%

How do I calculate statistical difference in above two and decide if its statistically significant?

I can do it by calculating yes and no response number from percentages given and do chi square test but it wont give a robust results.

Really appreciate input and help.

Re: Need robust method to calculate statistical difference

Reeza — Fri, 12 Jan 2018 15:59:06 GMT

Please do not post the same question multiple times. Note that I've merged these into a single post and moved them to the Statistical Procedures forum as that seems better suited.

Re: Need robust method to calculate statistical difference

StatDave — Fri, 12 Jan 2018 16:16:54 GMT

See the SURVEYFREQ procedure in SAS/STAT software.

Re: Need robust method to calculate statistical difference

GARYV — Fri, 12 Jan 2018 16:42:56 GMT

HI , Thank you for reply. I only have output table with percentages and C.I.. I don't have all the raw data set to use SURVEYFREQ procedure.

Re: Need robust method to calculate statistical difference

Reeza — Fri, 12 Jan 2018 17:29:18 GMT

All you need is the counts or you could recreate the data in an approximate fashion if you have the 2x2 numbers.

@GARYV wrote:

HI , Thank you for reply. I only have output table with percentages and C.I.. I don't have all the raw data set to use SURVEYFREQ procedure.

Re: Need robust method to calculate statistical difference

GARYV — Mon, 02 Apr 2018 19:39:46 GMT

You don't need to perform any statistical calculations. As long as CIs don't overlap, you can conclude that the results are statistically different. In the example in question, the result is statistically significant as CIs don't overlap.That is how governments interpret data. Correct me if I am wrong