Frodo Jedi
2018-Nov-12 01:07 UTC
[R] Reporting binomial logistic regression from R results
Dear list members, I need some help in understanding whether I am doing correctly a binomial logistic regression and whether I am interpreting the results in the correct way. Also I would need an advice regarding the reporting of the results from the R functions. I want to report the results of a binomial logistic regression where I want to assess difference between the 3 levels of a factor (called System) on the dependent variable (called Response) taking two values, 0 and 1. My goal is to understand if the effect of the 3 systems (A,B,C) in System affect differently Response in a significant way. I am basing my analysis on this URL: https://stats.idre.ucla.edu/r/dae/logit-regression/ This is the result of my analysis:> fit <- glm(Response ~ System, data = scrd, family = "binomial") > summary(fit)Call: glm(formula = Response ~ System, family = "binomial", data = scrd) Deviance Residuals: Min 1Q Median 3Q Max -2.8840 0.1775 0.2712 0.2712 0.5008 Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) 3.2844 0.2825 11.626 < 2e-16 *** SystemB -1.2715 0.3379 -3.763 0.000168 *** SystemC 0.8588 0.4990 1.721 0.085266 . --- Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 (Dispersion parameter for binomial family taken to be 1) Null deviance: 411.26 on 1023 degrees of freedom Residual deviance: 376.76 on 1021 degrees of freedom AIC: 382.76 Number of Fisher Scoring iterations: 6 Following this analysis I perform the wald test in order to understand whether there is an overall effect of System: library(aod)> wald.test(b = coef(fit), Sigma = vcov(fit), Terms = 1:3)Wald test: ---------- Chi-squared test: X2 = 354.6, df = 3, P(> X2) = 0.0 The chi-squared test statistic of 354.6, with 3 degrees of freedom is associated with a p-value < 0.001 indicating that the overall effect of System is statistically significant. Now I check whether there are differences between the coefficients using again the wald test: # Here difference between system B and C:> l <- cbind(0, 1, -1) > wald.test(b = coef(fit), Sigma = vcov(fit), L = l)Wald test: ---------- Chi-squared test: X2 = 22.3, df = 1, P(> X2) = 2.3e-06 # Here difference between system A and C:> l <- cbind(1, 0, -1) > wald.test(b = coef(fit), Sigma = vcov(fit), L = l)Wald test: ---------- Chi-squared test: X2 = 12.0, df = 1, P(> X2) = 0.00052 # Here difference between system A and B:> l <- cbind(1, -1, 0) > wald.test(b = coef(fit), Sigma = vcov(fit), L = l)Wald test: ---------- Chi-squared test: X2 = 58.7, df = 1, P(> X2) = 1.8e-14 My understanding is that from this analysis I can state that the three systems lead to a significantly different Response. Am I right? If so, how should I report the results of this analysis? What is the correct way? Thanks in advance Best wishes FJ [[alternative HTML version deleted]]
PIKAL Petr
2018-Nov-12 10:05 UTC
[R] Reporting binomial logistic regression from R results
Dear Frodo (or Jedi) The results seems to confirm your assumption that 3 systems are different. How you should present results probably depends on how it is usual to report such results in your environment. BTW. It seems to me like homework and this list has no homework policy (Sorry, if I am mistaken). Cheers Petr> -----Original Message----- > From: R-help <r-help-bounces at r-project.org> On Behalf Of Frodo Jedi > Sent: Monday, November 12, 2018 2:08 AM > To: r-help at r-project.org > Subject: [R] Reporting binomial logistic regression from R results > > Dear list members, > I need some help in understanding whether I am doing correctly a binomial > logistic regression and whether I am interpreting the results in the correct way. > Also I would need an advice regarding the reporting of the results from the R > functions. > > I want to report the results of a binomial logistic regression where I want to > assess difference between the 3 levels of a factor (called System) on the > dependent variable (called Response) taking two values, 0 and 1. My goal is to > understand if the effect of the 3 systems (A,B,C) in System affect differently > Response in a significant way. I am basing my analysis on this URL: > https://stats.idre.ucla.edu/r/dae/logit-regression/ > > This is the result of my analysis: > > > fit <- glm(Response ~ System, data = scrd, family = "binomial") > > summary(fit) > > Call: > glm(formula = Response ~ System, family = "binomial", data = scrd) > > Deviance Residuals: > Min 1Q Median 3Q Max > -2.8840 0.1775 0.2712 0.2712 0.5008 > > Coefficients: > Estimate Std. Error z value Pr(>|z|) > (Intercept) 3.2844 0.2825 11.626 < 2e-16 *** > SystemB -1.2715 0.3379 -3.763 0.000168 *** > SystemC 0.8588 0.4990 1.721 0.085266 . > --- > Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 > > (Dispersion parameter for binomial family taken to be 1) > > Null deviance: 411.26 on 1023 degrees of freedom Residual deviance: > 376.76 on 1021 degrees of freedom > AIC: 382.76 > > Number of Fisher Scoring iterations: 6 > Following this analysis I perform the wald test in order to understand whether > there is an overall effect of System: > > library(aod) > > > wald.test(b = coef(fit), Sigma = vcov(fit), Terms = 1:3) > Wald test: > ---------- > > Chi-squared test: > X2 = 354.6, df = 3, P(> X2) = 0.0 > The chi-squared test statistic of 354.6, with 3 degrees of freedom is associated > with a p-value < 0.001 indicating that the overall effect of System is statistically > significant. > > Now I check whether there are differences between the coefficients using again > the wald test: > > # Here difference between system B and C: > > > l <- cbind(0, 1, -1) > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 22.3, df = 1, P(> X2) = 2.3e-06 > > > > # Here difference between system A and C: > > > l <- cbind(1, 0, -1) > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 12.0, df = 1, P(> X2) = 0.00052 > > > > # Here difference between system A and B: > > > l <- cbind(1, -1, 0) > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 58.7, df = 1, P(> X2) = 1.8e-14 > > My understanding is that from this analysis I can state that the three systems > lead to a significantly different Response. Am I right? If so, how should I report > the results of this analysis? What is the correct way? > > > Thanks in advance > > Best wishes > > FJ > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.Osobn? ?daje: Informace o zpracov?n? a ochran? osobn?ch ?daj? obchodn?ch partner? PRECHEZA a.s. jsou zve?ejn?ny na: https://www.precheza.cz/zasady-ochrany-osobnich-udaju/ | Information about processing and protection of business partner?s personal data are available on website: https://www.precheza.cz/en/personal-data-protection-principles/ D?v?rnost: Tento e-mail a jak?koliv k n?mu p?ipojen? dokumenty jsou d?v?rn? a podl?haj? tomuto pr?vn? z?vazn?mu prohl??en? o vylou?en? odpov?dnosti: https://www.precheza.cz/01-dovetek/ | This email and any documents attached to it may be confidential and are subject to the legally binding disclaimer: https://www.precheza.cz/en/01-disclaimer/
PIKAL Petr
2018-Nov-12 13:02 UTC
[R] Reporting binomial logistic regression from R results
Hi Frodo I do not consider myself as an arbiter in statistical results and their presentation. Again your text seems to as good as any other. You should keep responses to mailing list as others could have another opinion. Cheers Petr From: Frodo Jedi <frodojedi.mailinglist at gmail.com> Sent: Monday, November 12, 2018 1:48 PM To: PIKAL Petr <petr.pikal at precheza.cz> Subject: Re: [R] Reporting binomial logistic regression from R results Dear Petr, many thanks for your reply. I was wondering whether in your opinion it is correct to report in a journal the following text: ?A logistic regression was performed to ascertain the effects of the system type on the likelihood that participants report correct identifications. The logistic regression model was statistically significant, ?2(3) = 354.6, p < 0.001, indicating an overall effect of the system type on participants' identification performances. The Wald test was used to compare the model coefficients related to the three systems. Results showed that participants? accuracy was significantly lower for the system B compared to both the system C (?2(1) = 22.3, p < 0.001) and the system A (?2(1) = 58.7, p < 0.001), as well as that the system C led to significantly higher identification accuracies than the system A (?2(1) = 12, p < 0.001).? Best wishes FJ On Mon, Nov 12, 2018 at 10:05 AM PIKAL Petr <petr.pikal at precheza.cz<mailto:petr.pikal at precheza.cz>> wrote: Dear Frodo (or Jedi) The results seems to confirm your assumption that 3 systems are different. How you should present results probably depends on how it is usual to report such results in your environment. BTW. It seems to me like homework and this list has no homework policy (Sorry, if I am mistaken). Cheers Petr> -----Original Message----- > From: R-help <r-help-bounces at r-project.org<mailto:r-help-bounces at r-project.org>> On Behalf Of Frodo Jedi > Sent: Monday, November 12, 2018 2:08 AM > To: r-help at r-project.org<mailto:r-help at r-project.org> > Subject: [R] Reporting binomial logistic regression from R results > > Dear list members, > I need some help in understanding whether I am doing correctly a binomial > logistic regression and whether I am interpreting the results in the correct way. > Also I would need an advice regarding the reporting of the results from the R > functions. > > I want to report the results of a binomial logistic regression where I want to > assess difference between the 3 levels of a factor (called System) on the > dependent variable (called Response) taking two values, 0 and 1. My goal is to > understand if the effect of the 3 systems (A,B,C) in System affect differently > Response in a significant way. I am basing my analysis on this URL: > https://stats.idre.ucla.edu/r/dae/logit-regression/ > > This is the result of my analysis: > > > fit <- glm(Response ~ System, data = scrd, family = "binomial") > > summary(fit) > > Call: > glm(formula = Response ~ System, family = "binomial", data = scrd) > > Deviance Residuals: > Min 1Q Median 3Q Max > -2.8840 0.1775 0.2712 0.2712 0.5008 > > Coefficients: > Estimate Std. Error z value Pr(>|z|) > (Intercept) 3.2844 0.2825 11.626 < 2e-16 *** > SystemB -1.2715 0.3379 -3.763 0.000168 *** > SystemC 0.8588 0.4990 1.721 0.085266 . > --- > Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 > > (Dispersion parameter for binomial family taken to be 1) > > Null deviance: 411.26 on 1023 degrees of freedom Residual deviance: > 376.76 on 1021 degrees of freedom > AIC: 382.76 > > Number of Fisher Scoring iterations: 6 > Following this analysis I perform the wald test in order to understand whether > there is an overall effect of System: > > library(aod) > > > wald.test(b = coef(fit), Sigma = vcov(fit), Terms = 1:3) > Wald test: > ---------- > > Chi-squared test: > X2 = 354.6, df = 3, P(> X2) = 0.0 > The chi-squared test statistic of 354.6, with 3 degrees of freedom is associated > with a p-value < 0.001 indicating that the overall effect of System is statistically > significant. > > Now I check whether there are differences between the coefficients using again > the wald test: > > # Here difference between system B and C: > > > l <- cbind(0, 1, -1) > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 22.3, df = 1, P(> X2) = 2.3e-06 > > > > # Here difference between system A and C: > > > l <- cbind(1, 0, -1) > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 12.0, df = 1, P(> X2) = 0.00052 > > > > # Here difference between system A and B: > > > l <- cbind(1, -1, 0) > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 58.7, df = 1, P(> X2) = 1.8e-14 > > My understanding is that from this analysis I can state that the three systems > lead to a significantly different Response. Am I right? If so, how should I report > the results of this analysis? What is the correct way? > > > Thanks in advance > > Best wishes > > FJ > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org<mailto:R-help at r-project.org> mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.Osobn? ?daje: Informace o zpracov?n? a ochran? osobn?ch ?daj? obchodn?ch partner? PRECHEZA a.s. jsou zve?ejn?ny na: https://www.precheza.cz/zasady-ochrany-osobnich-udaju/ | Information about processing and protection of business partner?s personal data are available on website: https://www.precheza.cz/en/personal-data-protection-principles/ D?v?rnost: Tento e-mail a jak?koliv k n?mu p?ipojen? dokumenty jsou d?v?rn? a podl?haj? tomuto pr?vn? z?vazn?mu prohl??en? o vylou?en? odpov?dnosti: https://www.precheza.cz/01-dovetek/ | This email and any documents attached to it may be confidential and are subject to the legally binding disclaimer: https://www.precheza.cz/en/01-disclaimer/ [[alternative HTML version deleted]]
Frodo Jedi
2018-Nov-12 13:06 UTC
[R] Reporting binomial logistic regression from R results
Dear Petr, thank you very much for your feedback. Can anyone in the list advise me if the way I report the results is correct? Kind regards FJ On Mon, Nov 12, 2018 at 1:02 PM PIKAL Petr <petr.pikal at precheza.cz> wrote:> Hi Frodo > > > > I do not consider myself as an arbiter in statistical results and their > presentation. Again your text seems to as good as any other. > > > > You should keep responses to mailing list as others could have another > opinion. > > > > Cheers > > Petr > > > > > > *From:* Frodo Jedi <frodojedi.mailinglist at gmail.com> > *Sent:* Monday, November 12, 2018 1:48 PM > *To:* PIKAL Petr <petr.pikal at precheza.cz> > *Subject:* Re: [R] Reporting binomial logistic regression from R results > > > > Dear Petr, > > many thanks for your reply. I was wondering whether in your opinion it is > correct to report in a journal the following text: > > > > > > ?A logistic regression was performed to ascertain the effects of the > system type on the likelihood that participants report correct > identifications. The logistic regression model was statistically > significant, ?2(3) = 354.6, p < 0.001, indicating an overall effect of the > system type on participants' identification performances. The Wald test was > used to compare the model coefficients related to the three systems. > Results showed that participants? accuracy was significantly lower for the > system B compared to both the system C (?2(1) = 22.3, p < 0.001) and the > system A (?2(1) = 58.7, p < 0.001), as well as that the system C led to > significantly higher identification accuracies than the system A (?2(1) > 12, p < 0.001).? > > > > > > Best wishes > > > > FJ > > > > > > > > > > > > On Mon, Nov 12, 2018 at 10:05 AM PIKAL Petr <petr.pikal at precheza.cz> > wrote: > > Dear Frodo (or Jedi) > > The results seems to confirm your assumption that 3 systems are different. > How you should present results probably depends on how it is usual to > report such results in your environment. > > BTW. It seems to me like homework and this list has no homework policy > (Sorry, if I am mistaken). > > Cheers > Petr > > -----Original Message----- > > From: R-help <r-help-bounces at r-project.org> On Behalf Of Frodo Jedi > > Sent: Monday, November 12, 2018 2:08 AM > > To: r-help at r-project.org > > Subject: [R] Reporting binomial logistic regression from R results > > > > Dear list members, > > I need some help in understanding whether I am doing correctly a binomial > > logistic regression and whether I am interpreting the results in the > correct way. > > Also I would need an advice regarding the reporting of the results from > the R > > functions. > > > > I want to report the results of a binomial logistic regression where I > want to > > assess difference between the 3 levels of a factor (called System) on the > > dependent variable (called Response) taking two values, 0 and 1. My goal > is to > > understand if the effect of the 3 systems (A,B,C) in System affect > differently > > Response in a significant way. I am basing my analysis on this URL: > > https://stats.idre.ucla.edu/r/dae/logit-regression/ > > > > This is the result of my analysis: > > > > > fit <- glm(Response ~ System, data = scrd, family = "binomial") > > > summary(fit) > > > > Call: > > glm(formula = Response ~ System, family = "binomial", data = scrd) > > > > Deviance Residuals: > > Min 1Q Median 3Q Max > > -2.8840 0.1775 0.2712 0.2712 0.5008 > > > > Coefficients: > > Estimate Std. Error z value Pr(>|z|) > > (Intercept) 3.2844 0.2825 11.626 < 2e-16 *** > > SystemB -1.2715 0.3379 -3.763 0.000168 *** > > SystemC 0.8588 0.4990 1.721 0.085266 . > > --- > > Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 > > > > (Dispersion parameter for binomial family taken to be 1) > > > > Null deviance: 411.26 on 1023 degrees of freedom Residual deviance: > > 376.76 on 1021 degrees of freedom > > AIC: 382.76 > > > > Number of Fisher Scoring iterations: 6 > > Following this analysis I perform the wald test in order to understand > whether > > there is an overall effect of System: > > > > library(aod) > > > > > wald.test(b = coef(fit), Sigma = vcov(fit), Terms = 1:3) > > Wald test: > > ---------- > > > > Chi-squared test: > > X2 = 354.6, df = 3, P(> X2) = 0.0 > > The chi-squared test statistic of 354.6, with 3 degrees of freedom is > associated > > with a p-value < 0.001 indicating that the overall effect of System is > statistically > > significant. > > > > Now I check whether there are differences between the coefficients using > again > > the wald test: > > > > # Here difference between system B and C: > > > > > l <- cbind(0, 1, -1) > > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > > Wald test: > > ---------- > > > > Chi-squared test: > > X2 = 22.3, df = 1, P(> X2) = 2.3e-06 > > > > > > > > # Here difference between system A and C: > > > > > l <- cbind(1, 0, -1) > > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > > Wald test: > > ---------- > > > > Chi-squared test: > > X2 = 12.0, df = 1, P(> X2) = 0.00052 > > > > > > > > # Here difference between system A and B: > > > > > l <- cbind(1, -1, 0) > > > wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > > Wald test: > > ---------- > > > > Chi-squared test: > > X2 = 58.7, df = 1, P(> X2) = 1.8e-14 > > > > My understanding is that from this analysis I can state that the three > systems > > lead to a significantly different Response. Am I right? If so, how > should I report > > the results of this analysis? What is the correct way? > > > > > > Thanks in advance > > > > Best wishes > > > > FJ > > > > [[alternative HTML version deleted]] > > > > ______________________________________________ > > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > > and provide commented, minimal, self-contained, reproducible code. > Osobn? ?daje: Informace o zpracov?n? a ochran? osobn?ch ?daj? obchodn?ch > partner? PRECHEZA a.s. jsou zve?ejn?ny na: > https://www.precheza.cz/zasady-ochrany-osobnich-udaju/ | Information > about processing and protection of business partner?s personal data are > available on website: > https://www.precheza.cz/en/personal-data-protection-principles/ > D?v?rnost: Tento e-mail a jak?koliv k n?mu p?ipojen? dokumenty jsou > d?v?rn? a podl?haj? tomuto pr?vn? z?vazn?mu prohl??en? o vylou?en? > odpov?dnosti: https://www.precheza.cz/01-dovetek/ | This email and any > documents attached to it may be confidential and are subject to the legally > binding disclaimer: https://www.precheza.cz/en/01-disclaimer/ > >[[alternative HTML version deleted]]
Eik Vettorazzi
2018-Nov-12 13:15 UTC
[R] Reporting binomial logistic regression from R results
Dear Jedi, please use the source carefully. A and C are not statistically different at the 5% level, which can be inferred from glm output. Your last two wald.tests don't test what you want to, since your model contains an intercept term. You specified contrasts which tests A vs B-A, ie A- (B-A)==0 <-> 2*A-B==0 which is not intended I think. Have a look at ?contr.treatment and re-read your source doc to get an idea what dummy coding and indicatr variables are about. Cheers Am 12.11.2018 um 02:07 schrieb Frodo Jedi:> Dear list members, > I need some help in understanding whether I am doing correctly a binomial > logistic regression and whether I am interpreting the results in the > correct way. Also I would need an advice regarding the reporting of the > results from the R functions. > > I want to report the results of a binomial logistic regression where I want > to assess difference between the 3 levels of a factor (called System) on > the dependent variable (called Response) taking two values, 0 and 1. My > goal is to understand if the effect of the 3 systems (A,B,C) in System > affect differently Response in a significant way. I am basing my analysis > on this URL: https://stats.idre.ucla.edu/r/dae/logit-regression/ > > This is the result of my analysis: > >> fit <- glm(Response ~ System, data = scrd, family = "binomial") >> summary(fit) > > Call: > glm(formula = Response ~ System, family = "binomial", data = scrd) > > Deviance Residuals: > Min 1Q Median 3Q Max > -2.8840 0.1775 0.2712 0.2712 0.5008 > > Coefficients: > Estimate Std. Error z value Pr(>|z|) > (Intercept) 3.2844 0.2825 11.626 < 2e-16 *** > SystemB -1.2715 0.3379 -3.763 0.000168 *** > SystemC 0.8588 0.4990 1.721 0.085266 . > --- > Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 > > (Dispersion parameter for binomial family taken to be 1) > > Null deviance: 411.26 on 1023 degrees of freedom > Residual deviance: 376.76 on 1021 degrees of freedom > AIC: 382.76 > > Number of Fisher Scoring iterations: 6 > Following this analysis I perform the wald test in order to understand > whether there is an overall effect of System: > > library(aod) > >> wald.test(b = coef(fit), Sigma = vcov(fit), Terms = 1:3) > Wald test: > ---------- > > Chi-squared test: > X2 = 354.6, df = 3, P(> X2) = 0.0 > The chi-squared test statistic of 354.6, with 3 degrees of freedom is > associated with a p-value < 0.001 indicating that the overall effect of > System is statistically significant. > > Now I check whether there are differences between the coefficients using > again the wald test: > > # Here difference between system B and C: > >> l <- cbind(0, 1, -1) >> wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 22.3, df = 1, P(> X2) = 2.3e-06 > > > > # Here difference between system A and C: > >> l <- cbind(1, 0, -1) >> wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 12.0, df = 1, P(> X2) = 0.00052 > > > > # Here difference between system A and B: > >> l <- cbind(1, -1, 0) >> wald.test(b = coef(fit), Sigma = vcov(fit), L = l) > Wald test: > ---------- > > Chi-squared test: > X2 = 58.7, df = 1, P(> X2) = 1.8e-14 > > My understanding is that from this analysis I can state that the three > systems lead to a significantly different Response. Am I right? If so, how > should I report the results of this analysis? What is the correct way? > > > Thanks in advance > > Best wishes > > FJ > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >-- Eik Vettorazzi Department of Medical Biometry and Epidemiology University Medical Center Hamburg-Eppendorf Martinistrasse 52 building W 34 20246 Hamburg Phone: +49 (0) 40 7410 - 58243 Fax: +49 (0) 40 7410 - 57790 Web: www.uke.de/imbe -- _____________________________________________________________________ Universit?tsklinikum Hamburg-Eppendorf; K?rperschaft des ?ffentlichen Rechts; Gerichtsstand: Hamburg | www.uke.de Vorstandsmitglieder: Prof. Dr. Burkhard G?ke (Vorsitzender), Prof. Dr. Dr. Uwe Koch-Gromus, Joachim Pr?l?, Marya Verdel _____________________________________________________________________ SAVE PAPER - THINK BEFORE PRINTING
peter dalgaard
2018-Nov-12 13:46 UTC
[R] Reporting binomial logistic regression from R results
Yes, only one of the pairwise comparisons (B vs. C) is right. Also, the overall test has 3 degrees of freedom whereas a comparison of 3 groups should have 2. You (meaning Frodo) are testing that _all 3_ regression coefficients are zero, intercept included. That would imply that all three systems have response probablilities og 0.5, which is not likely what you want. This all suggests that you are struggling with the interpretation of the regression coefficients and their role in the linear predictor. This should be covered by any good book on logistic regression. -pd> On 12 Nov 2018, at 14:15 , Eik Vettorazzi <E.Vettorazzi at uke.de> wrote: > > Dear Jedi, > please use the source carefully. A and C are not statistically different at the 5% level, which can be inferred from glm output. Your last two wald.tests don't test what you want to, since your model contains an intercept term. You specified contrasts which tests A vs B-A, ie A- (B-A)==0 <-> 2*A-B==0 which is not intended I think. Have a look at ?contr.treatment and re-read your source doc to get an idea what dummy coding and indicatr variables are about. > > Cheers > > > Am 12.11.2018 um 02:07 schrieb Frodo Jedi: >> Dear list members, >> I need some help in understanding whether I am doing correctly a binomial >> logistic regression and whether I am interpreting the results in the >> correct way. Also I would need an advice regarding the reporting of the >> results from the R functions. >> I want to report the results of a binomial logistic regression where I want >> to assess difference between the 3 levels of a factor (called System) on >> the dependent variable (called Response) taking two values, 0 and 1. My >> goal is to understand if the effect of the 3 systems (A,B,C) in System >> affect differently Response in a significant way. I am basing my analysis >> on this URL: https://stats.idre.ucla.edu/r/dae/logit-regression/ >> This is the result of my analysis: >>> fit <- glm(Response ~ System, data = scrd, family = "binomial") >>> summary(fit) >> Call: >> glm(formula = Response ~ System, family = "binomial", data = scrd) >> Deviance Residuals: >> Min 1Q Median 3Q Max >> -2.8840 0.1775 0.2712 0.2712 0.5008 >> Coefficients: >> Estimate Std. Error z value Pr(>|z|) >> (Intercept) 3.2844 0.2825 11.626 < 2e-16 *** >> SystemB -1.2715 0.3379 -3.763 0.000168 *** >> SystemC 0.8588 0.4990 1.721 0.085266 . >> --- >> Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 >> (Dispersion parameter for binomial family taken to be 1) >> Null deviance: 411.26 on 1023 degrees of freedom >> Residual deviance: 376.76 on 1021 degrees of freedom >> AIC: 382.76 >> Number of Fisher Scoring iterations: 6 >> Following this analysis I perform the wald test in order to understand >> whether there is an overall effect of System: >> library(aod) >>> wald.test(b = coef(fit), Sigma = vcov(fit), Terms = 1:3) >> Wald test: >> ---------- >> Chi-squared test: >> X2 = 354.6, df = 3, P(> X2) = 0.0 >> The chi-squared test statistic of 354.6, with 3 degrees of freedom is >> associated with a p-value < 0.001 indicating that the overall effect of >> System is statistically significant. >> Now I check whether there are differences between the coefficients using >> again the wald test: >> # Here difference between system B and C: >>> l <- cbind(0, 1, -1) >>> wald.test(b = coef(fit), Sigma = vcov(fit), L = l) >> Wald test: >> ---------- >> Chi-squared test: >> X2 = 22.3, df = 1, P(> X2) = 2.3e-06 >> # Here difference between system A and C: >>> l <- cbind(1, 0, -1) >>> wald.test(b = coef(fit), Sigma = vcov(fit), L = l) >> Wald test: >> ---------- >> Chi-squared test: >> X2 = 12.0, df = 1, P(> X2) = 0.00052 >> # Here difference between system A and B: >>> l <- cbind(1, -1, 0) >>> wald.test(b = coef(fit), Sigma = vcov(fit), L = l) >> Wald test: >> ---------- >> Chi-squared test: >> X2 = 58.7, df = 1, P(> X2) = 1.8e-14 >> My understanding is that from this analysis I can state that the three >> systems lead to a significantly different Response. Am I right? If so, how >> should I report the results of this analysis? What is the correct way? >> Thanks in advance >> Best wishes >> FJ >> [[alternative HTML version deleted]] >> ______________________________________________ >> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. > > -- > Eik Vettorazzi > > Department of Medical Biometry and Epidemiology > University Medical Center Hamburg-Eppendorf > > Martinistrasse 52 > building W 34 > 20246 Hamburg > > Phone: +49 (0) 40 7410 - 58243 > Fax: +49 (0) 40 7410 - 57790 > Web: www.uke.de/imbe > -- > > _____________________________________________________________________ > > Universit?tsklinikum Hamburg-Eppendorf; K?rperschaft des ?ffentlichen Rechts; Gerichtsstand: Hamburg | www.uke.de > Vorstandsmitglieder: Prof. Dr. Burkhard G?ke (Vorsitzender), Prof. Dr. Dr. Uwe Koch-Gromus, Joachim Pr?l?, Marya Verdel > _____________________________________________________________________ > > SAVE PAPER - THINK BEFORE PRINTING > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.-- Peter Dalgaard, Professor, Center for Statistics, Copenhagen Business School Solbjerg Plads 3, 2000 Frederiksberg, Denmark Phone: (+45)38153501 Office: A 4.23 Email: pd.mes at cbs.dk Priv: PDalgd at gmail.com