Sig T on Computed Category

[Arora, Manoj (IMDLR)]

Hi All,

 

Please find below the table in which there is a One row which is calculated using PCOMPUTE command . If I calculate sig t it doesnot work on computed category. Please suggest any script or python utility to calculate the same

 

 

BASE : All Total		Fusion_method 11 clusters final-Nov30
		Total	CL1	CL2	CL3	CL4	CL5	CL6	CL7	CL8	CL9	CL10	CL11
Q39	Answering Base	17839	1419	1999	1586	1590	1678	1652	1768	1154	1907	1457	1628
	Strongly Disagree	.7	.8	1.0	1.4	3.1	.3	.5	.2	.1	.1	.3	.3
	2	5.1	8.3	6.6	7.6	19.1	3.8	3.0	2.5	2.5	.9	1.5	.5
	3	3.0	5.1	6.1	3.1	7.0	3.2	2.6	1.7	.2	1.0	1.4	.6
	4	4.7	6.6	8.3	5.9	8.1	3.9	3.6	3.7	1.0	3.5	3.6	2.2
	5	12.8	15.4	17.3	17.1	16.9	10.3	11.9	10.9	5.4	13.2	10.1	9.2
	6	17.7	20.3	22.0	18.2	17.9	15.0	17.6	16.4	10.6	19.9	16.6	17.4
	7	18.1	16.2	15.2	14.5	9.4	16.4	19.5	19.8	13.7	24.4	22.1	26.1
	8	17.0	14.0	12.3	12.8	6.8	16.3	18.4	20.9	20.8	20.3	20.0	25.8
	9	11.5	6.7	6.7	8.6	5.6	16.2	13.4	12.7	19.7	10.8	16.2	12.5
	Strongly Agree	9.4	6.7	4.6	10.8	6.2	14.6	9.5	11.2	26.1	5.9	8.2	5.5
	Detractor	5.8%	9.2%	7.6%	9.0%	22.2%	4.1%	3.5%	2.7%	2.5%	1.0%	1.8%	.8%
	Passive	20.9%	13.3%	11.3%	19.4%	11.8%	30.8%	22.9%	23.9%	45.8%	16.7%	24.4%	18.0%
	Promoters	73.3%	77.5%	81.1%	71.6%	66.1%	65.1%	73.6%	73.4%	51.7%	82.3%	73.9%	81.2%
	NPS	67.5%	68.4%	73.4%	62.6%	43.9%	61.0%	70.1%	70.6%	49.1%	81.4%	72.1%	80.4%
Mean		6.69	6.10	5.95	6.33	5.19	7.12	6.93	7.10	8.01	6.94	7.15	7.18
Standard Deviation		2.12	2.19	2.08	2.30	2.48	2.14	1.98	1.90	1.85	1.62	1.78	1.51

 

 

Want to calculate the Sig T on NPS row which is generated using PCOMPUTE command.

 

Regards

Manoj

 

Kantar Disclaimer ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Jon Peck]

If I understand what you want, it would be a column proportions test, not a t test.  However, since a PCOMPUTE is a function of the other categories in the column, it would not be independent of them, although in this table, it is hard to tell what the subtable consists of.  It looks like there is only one category.  Nonindepence means that the test calculation would not be valid, which is why it is left out.

On Fri, Dec 28, 2018 at 3:31 AM Arora, Manoj (IMDLR) <[hidden email]> wrote:

Hi All,

 

Please find below the table in which there is a One row which is calculated using PCOMPUTE command . If I calculate sig t it doesnot work on computed category. Please suggest any script or python utility to calculate the same

 

 

BASE : All Total		Fusion_method 11 clusters final-Nov30
		Total	CL1	CL2	CL3	CL4	CL5	CL6	CL7	CL8	CL9	CL10	CL11
Q39	Answering Base	17839	1419	1999	1586	1590	1678	1652	1768	1154	1907	1457	1628
	Strongly Disagree	.7	.8	1.0	1.4	3.1	.3	.5	.2	.1	.1	.3	.3
	2	5.1	8.3	6.6	7.6	19.1	3.8	3.0	2.5	2.5	.9	1.5	.5
	3	3.0	5.1	6.1	3.1	7.0	3.2	2.6	1.7	.2	1.0	1.4	.6
	4	4.7	6.6	8.3	5.9	8.1	3.9	3.6	3.7	1.0	3.5	3.6	2.2
	5	12.8	15.4	17.3	17.1	16.9	10.3	11.9	10.9	5.4	13.2	10.1	9.2
	6	17.7	20.3	22.0	18.2	17.9	15.0	17.6	16.4	10.6	19.9	16.6	17.4
	7	18.1	16.2	15.2	14.5	9.4	16.4	19.5	19.8	13.7	24.4	22.1	26.1
	8	17.0	14.0	12.3	12.8	6.8	16.3	18.4	20.9	20.8	20.3	20.0	25.8
	9	11.5	6.7	6.7	8.6	5.6	16.2	13.4	12.7	19.7	10.8	16.2	12.5
	Strongly Agree	9.4	6.7	4.6	10.8	6.2	14.6	9.5	11.2	26.1	5.9	8.2	5.5
	Detractor	5.8%	9.2%	7.6%	9.0%	22.2%	4.1%	3.5%	2.7%	2.5%	1.0%	1.8%	.8%
	Passive	20.9%	13.3%	11.3%	19.4%	11.8%	30.8%	22.9%	23.9%	45.8%	16.7%	24.4%	18.0%
	Promoters	73.3%	77.5%	81.1%	71.6%	66.1%	65.1%	73.6%	73.4%	51.7%	82.3%	73.9%	81.2%
	NPS	67.5%	68.4%	73.4%	62.6%	43.9%	61.0%	70.1%	70.6%	49.1%	81.4%	72.1%	80.4%
Mean		6.69	6.10	5.95	6.33	5.19	7.12	6.93	7.10	8.01	6.94	7.15	7.18
Standard Deviation		2.12	2.19	2.08	2.30	2.48	2.14	1.98	1.90	1.85	1.62	1.78	1.51

 

 

Want to calculate the Sig T on NPS row which is generated using PCOMPUTE command.

 

Regards

Manoj

 

Kantar Disclaimer

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

--
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD