proportions: (a,b,c)/3 vs d?
Locked
 
|
I have 4 proportions (N=111): a=.117 (n=13); b=.018(n=2); c=.099(n=11); and d(n=85)=.766. I want to know if d is significantly different from the average of a,b, and c. I don't have access to Marascuilo & McSweeney's book. I got Cohen 1967 article: An alternative to Marascuilo's "Large-sample multiple comparisons" for proportions. But, it involves an arcsin transformation of p. Is there a simple way to do this contrast using SPSS? I've got 36 of these contrasts to do.
z = ((a)+(b)+(c))/3 -1.0(d)/SE(?) Any suggestions will be greatly appreciated. TIA Stephen Salbod, Pace University, NYC |
|
How did you get a proortion of .018 with an n of 2?
0/2 =0, 1/2=.5 2/2=1.0
Art Kendall
Social Research Consultants |
|
|
Art, It is the proportion which is 0.18 Frequencies could be 10 and 550 which yield proportion 0.018 ... Mark Miller On Wed, Nov 5, 2014 at 8:58 AM, Art Kendall <[hidden email]> wrote: How did you get a proortion of .018 with an n of 2? |
|
|
if the denominator were 550 then the n would be 550. The OP said the n was 2.
Or am I missing something? Is it in fact a 2-way design?
Art Kendall
Social Research Consultants |
|
|
In reply to this post by Art Kendall
Hi Art, 2/111 = .018.
I might not be asking the right question. One hundred and eleven students heard an actor give a video message and selected the emotional tone of the message. The message was a single sentence. The actor had been given the intent message. I want to test whether the IV was working. I thought a contrast within each message should do the trick? Is there another way to skin the cat? --Steve -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall Sent: Wednesday, November 05, 2014 11:59 AM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? How did you get a proortion of .018 with an n of 2? 0/2 =0, 1/2=.5 2/2=1.0 ----- Art Kendall Social Research Consultants -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/proportions-a-b-c-3-vs-d-tp5727795p5727797.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== 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 |
Re: proportions: (a,b,c)/3 vs d?
Administrator
|
In reply to this post by Art Kendall
From the OP:
"I have 4 proportions (N=111): a=.117 (n=13); b=.018(n=2); c=.099(n=11); and d(n=85)=.766." My interpretation: a. 13/111 = 0.117117117 b. 2/111 = 0.018018018 c. 11/111 = 0.099099099 d. 85/111 = 0.765765766
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
In reply to this post by Art Kendall
Hi Art,
You're right; I did not express the problem correctly. It is a one sample problem. -Steve -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall Sent: Wednesday, November 05, 2014 1:05 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? if the denominator were 550 then the n would be 550. The OP said the n was 2. Or am I missing something? Is it in fact a 2-way design? ----- Art Kendall Social Research Consultants -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/proportions-a-b-c-3-vs-d-tp5727795p5727800.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== 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 |
Re: proportions: (a,b,c)/3 vs d?
|
|
Complete enough information is always better. So was it that students saw/heard the actor deliver the line and then marked which of four options (e.g., 'angry', 'sad', 'joyous' 'disgusted') best described what they saw/heard? Or something else?
Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Salbod, Mr. Stephen Sent: Wednesday, November 05, 2014 4:02 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? Hi Art, You're right; I did not express the problem correctly. It is a one sample problem. -Steve -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall Sent: Wednesday, November 05, 2014 1:05 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? if the denominator were 550 then the n would be 550. The OP said the n was 2. Or am I missing something? Is it in fact a 2-way design? ----- Art Kendall Social Research Consultants -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/proportions-a-b-c-3-vs-d-tp5727795p5727800.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== 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 ===================== 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 |
Re: proportions: (a,b,c)/3 vs d?
|
Administrator
|
In reply to this post by Salbod
It sounds to me like you want a chi-square goodness of fit (GOF) test. But I cannot yet work out what your expected frequencies are under the null hypothesis. IF the null stated equal proportions in the 4 categories, it would be something like this, I think, with the second test using NewCat being the one that contrasts 1-3 vs 4.
DATA LIST LIST /Category Kount(2F5.0) . BEGIN DATA. 1 13 2 2 3 11 4 85 END DATA. WEIGHT by Kount. FREQUENCIES Category. * Chi-square GOF test with all 4 categories, * and a null hypothesis stating that 25% of * the observations fall in each category. NPAR TESTS /CHISQUARE=Category /EXPECTED= .25 .25 .25 .25 /STATISTICS DESCRIPTIVES /MISSING ANALYSIS. RECODE Category (1 2 3 = 1)(4=2) INTO NewCat. FORMATS NewCat(F1). FREQUENCIES NewCat. * Chi-square GOF test with catgegories 1-3 * combined vs category 4, and a null hypothesis * stating that 75% of the cases fall in categories * 1-3 and 25% fall in category 4. NPAR TESTS /CHISQUARE=NewCat /EXPECTED= .75 .25 /STATISTICS DESCRIPTIVES /MISSING ANALYSIS. But as I said, I'm not at all sure that your null specifies equal proportions in the 4 categories. HTH.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
Bruce, thank you for the code. It getting me to see where my problem is. The expected values are a problem because I've no idea what to expect. If everything worked perfectly, d would have n = 111: that is why I went with a focused contrast (df=1). I think it is boiling down to how to calculate the associated SE.
--Steve -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bruce Weaver Sent: Wednesday, November 05, 2014 4:40 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? It sounds to me like you want a chi-square goodness of fit (GOF) test. But I cannot yet work out what your expected frequencies are under the null hypothesis. IF the null stated equal proportions in the 4 categories, it would be something like this, I think, with the second test using NewCat being the one that contrasts 1-3 vs 4. DATA LIST LIST /Category Kount(2F5.0) . BEGIN DATA. 1 13 2 2 3 11 4 85 END DATA. WEIGHT by Kount. FREQUENCIES Category. * Chi-square GOF test with all 4 categories, * and a null hypothesis stating that 25% of * the observations fall in each category. NPAR TESTS /CHISQUARE=Category /EXPECTED= .25 .25 .25 .25 /STATISTICS DESCRIPTIVES /MISSING ANALYSIS. RECODE Category (1 2 3 = 1)(4=2) INTO NewCat. FORMATS NewCat(F1). FREQUENCIES NewCat. * Chi-square GOF test with catgegories 1-3 * combined vs category 4, and a null hypothesis * stating that 75% of the cases fall in categories * 1-3 and 25% fall in category 4. NPAR TESTS /CHISQUARE=NewCat /EXPECTED= .75 .25 /STATISTICS DESCRIPTIVES /MISSING ANALYSIS. But as I said, I'm not at all sure that /your/ null specifies equal proportions in the 4 categories. HTH. Salbod wrote > Hi Art, 2/111 = .018. > > I might not be asking the right question. One hundred and eleven > students heard an actor give a video message and selected the > emotional tone of the message. The message was a single sentence. The > actor had been given the intent message. I want to test whether the IV > was working. I thought a contrast within each message should do the > trick? Is there another way to skin the cat? > > --Steve > > -----Original Message----- > From: SPSSX(r) Discussion [mailto: > SPSSX-L@.UGA > ] On Behalf Of Art Kendall > Sent: Wednesday, November 05, 2014 11:59 AM > To: > SPSSX-L@.UGA > Subject: Re: proportions: (a,b,c)/3 vs d? > > How did you get a proortion of .018 with an n of 2? > 0/2 =0, 1/2=.5 2/2=1.0 > > > > ----- > Art Kendall > Social Research Consultants > -- > View this message in context: > http://spssx-discussion.1045642.n5.nabble.com/proportions-a-b-c-3-vs-d > -tp5727795p5727797.html Sent from the SPSSX Discussion mailing list > archive at Nabble.com. > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (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 > LISTSERV@.UGA > (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 ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/proportions-a-b-c-3-vs-d-tp5727795p5727810.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== 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 |
|
|
In reply to this post by Salbod
As I read this, for the statement of the problem as, "I want to test
=====================
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
whether the IV was working"... there is one correct answer for the particular message. Like on an IQ test, the important information is not a matter of which of the *wrong* answers was given. If you are doing scale development, you may care about which categories are confused most, when it comes to figuring out improvements, etc. This is hardly a situation for a *test*. You should not be happy with something that is a mere 0.05 or even 0.01. If your actor does not do a good job showing the intent -- something a lot better than chance -- then the item will not work for the future purposes. Is 76% a good fraction? Well, you can put a CI on it to see how wide it is. If you have a number of these as potential items, your best display might be to list them in increasing number of Wrongs. The Odds Ratio for 75% is 3.0, if you want another way to describe the effectiveness. -- Rich Ulrich > Date: Wed, 5 Nov 2014 20:57:09 +0000 > From: [hidden email] > Subject: Re: proportions: (a,b,c)/3 vs d? > To: [hidden email] > > Hi Art, 2/111 = .018. > > I might not be asking the right question. One hundred and eleven students heard an actor give a video message and selected the emotional tone of the message. The message was a single sentence. The actor had been given the intent message. I want to test whether the IV was working. I thought a contrast within each message should do the trick? Is there another way to skin the cat? > > --Steve |
|
|
Thanks for getting back to me and for taking the time to consider the problem. You’ve given me another lens to view my problem through. Let go back to the drawing
board. --Steve From: Rich Ulrich [mailto:[hidden email]]
As I read this, for the statement of the problem as, "I want to test
> Date: Wed, 5 Nov 2014 20:57:09 +0000 |
|
|
I'm sorry I didn't attach Cohen's article earlier. --SteveCohen1967.pdf
|
|
|
In reply to this post by Maguin, Eugene
Hi Gene,
I realized after considerably thought to everyone's responses, how off base my question was. It reminded me of my struggle with the Monty Hall problem. I have a one sample problem that I needed to model responses to. All my models I came up with (e.g., normal curve) produced significant chi-squares; therefore, I used the rubric that the message had a particular emotional tone if more than half the respondents selected that level. I got agreement between respondents and intent on 32 of 36 messages. Thanks everyone, Steve -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Maguin, Eugene Sent: Wednesday, November 05, 2014 4:24 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? Complete enough information is always better. So was it that students saw/heard the actor deliver the line and then marked which of four options (e.g., 'angry', 'sad', 'joyous' 'disgusted') best described what they saw/heard? Or something else? Gene Maguin -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Salbod, Mr. Stephen Sent: Wednesday, November 05, 2014 4:02 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? Hi Art, You're right; I did not express the problem correctly. It is a one sample problem. -Steve -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall Sent: Wednesday, November 05, 2014 1:05 PM To: [hidden email] Subject: Re: proportions: (a,b,c)/3 vs d? if the denominator were 550 then the n would be 550. The OP said the n was 2. Or am I missing something? Is it in fact a 2-way design? ----- Art Kendall Social Research Consultants -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/proportions-a-b-c-3-vs-d-tp5727795p5727800.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== 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 ===================== 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 |
|
|
In reply to this post by Salbod
Steve,
=====================
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
Whether or not such a test is meaningful has been rightfully called into question. Having said that, it is certainly possible to test that very hypothesis. Assuming you didn't have such sparse data (making it reasonable to employ maximum likelihood estimation of parameters), I would recommend you fit a multinomial logistic regression model without any predictors, and perform the contrast by estimating the SE via the delta method via the NLMIXED procedure in SAS. Here is the untested code: proc nlmixed data = multinomial_response; parms beta1_0 0 beta2_0 0 beta3_0 0; eta1 = beta1_0; eta2 = beta2_0; eta3 = beta3_0; prob1 = exp(eta1) / (1 + exp(eta1) + exp(eta2) + exp(eta3)); prob2 = exp(eta2) / (1 + exp(eta1) + exp(eta2) + exp(eta3)); prob3 = exp(eta3) / (1 + exp(eta1) + exp(eta2) + exp(eta3)); prob4 = 1 / (1 + exp(eta1) + exp(eta2) + exp(eta3)); if y = 1 then loglike = log(prob1); else if y = 2 then loglike = log(prob2); else if y = 3 then loglike = log(prob3); else loglike = log(prob4); model y ~ general(loglike); estimate 'contrast' (prob1 + prob2 + prob3)/3 - prob4; run; The dataset should be constructed as follows: ID y 1 1 2 1 3 2 4 1 5 3 6 2 7 4 8 4 9 3 . . . Ryan On Wed, Nov 5, 2014 at 6:35 PM, Salbod, Mr. Stephen <[hidden email]> wrote: Bruce, thank you for the code. It getting me to see where my problem is. The expected values are a problem because I've no idea what to expect. If everything worked perfectly, d would have n = 111: that is why I went with a focused contrast (df=1). I think it is boiling down to how to calculate the associated SE. |
|
|
Wow. Thank you, Ryan.
This is going to keep me busy for a long time
J
I love hard fun, Steve From: SPSSX(r) Discussion [mailto:[hidden email]]
On Behalf Of Ryan Black Steve, Whether or not such a test is meaningful has been rightfully called into question. Having said that, it is certainly possible to test that very hypothesis. Assuming you didn't have such sparse data (making it reasonable to employ maximum
likelihood estimation of parameters), I would recommend you fit a multinomial logistic regression model without any predictors, and perform the contrast by estimating the SE via the delta method via the NLMIXED procedure in SAS. Here is the untested code: proc nlmixed data = multinomial_response; parms beta1_0 0 beta2_0 0 beta3_0 0; eta1 = beta1_0; eta2 = beta2_0; eta3 = beta3_0; prob1 = exp(eta1) / (1 + exp(eta1) + exp(eta2) + exp(eta3)); prob2 = exp(eta2) / (1 + exp(eta1) + exp(eta2) + exp(eta3)); prob3 = exp(eta3) / (1 + exp(eta1) + exp(eta2) + exp(eta3)); prob4 = 1 / (1 + exp(eta1) + exp(eta2) + exp(eta3)); if y = 1 then loglike = log(prob1); else if y = 2 then loglike = log(prob2); else if y = 3 then loglike = log(prob3); else loglike = log(prob4); model y ~ general(loglike); estimate 'contrast' (prob1 + prob2 + prob3)/3 - prob4; run; The dataset should be constructed as follows: ID y 1 1 2 1 3 2 4 1 5 3 6 2 7 4 8 4 9 3 . . . Ryan On Wed, Nov 5, 2014 at 6:35 PM, Salbod, Mr. Stephen <[hidden email]> wrote:
===================== 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
|
| Free forum by Nabble | Edit this page |