Dear All, I am using the "effects" package to produce predicted probability from a logistic regression. The graph looks really good. I soon realized that the y-axis is not spaced equally. For example, in my case, the distance between 0.02 and 0.04 is much greater than that between 0.06 and 0.08. I can guess a reason for this, but unfortunately, this distorts my story. Is there a way to change this and make the y-axis equally spaced? Many thanks. Best, Shige
Dear Shige, As documented in ?plot.eff, the default is to plot on the scale of the linear predictor (the logit scale, for a logit model), which preserves the linearity of the model (which, I would think, is generally desirable), but to label the axis on the scale of the response (the probability scale). As is also documented, setting rescale.axis=FALSE will plot on the scale of the response. I hope this helps, John -------------------------------- John Fox Senator William McMaster Professor of Social Statistics Department of Sociology McMaster University Hamilton, Ontario, Canada web: socserv.mcmaster.ca/jfox> -----Original Message----- > From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]On> Behalf Of Shige Song > Sent: November-13-10 10:14 AM > To: r-help Help > Subject: [R] Question about the "effects" package > > Dear All, > > I am using the "effects" package to produce predicted probability from > a logistic regression. The graph looks really good. I soon realized > that the y-axis is not spaced equally. For example, in my case, the > distance between 0.02 and 0.04 is much greater than that between 0.06 > and 0.08. I can guess a reason for this, but unfortunately, this > distorts my story. Is there a way to change this and make the y-axis > equally spaced? > > Many thanks. > > Best, > Shige > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html> and provide commented, minimal, self-contained, reproducible code.
Dear John, This is exactly what I need, thank you so much for the tip (and thank you so much for this wonderful package too). Best, Shige On Sat, Nov 13, 2010 at 10:50 AM, John Fox <jfox at mcmaster.ca> wrote:> Dear Shige, > > As documented in ?plot.eff, the default is to plot on the scale of the > linear predictor (the logit scale, for a logit model), which preserves the > linearity of the model (which, I would think, is generally desirable), but > to label the axis on the scale of the response (the probability scale). As > is also documented, setting rescale.axis=FALSE will plot on the scale of the > response. > > I hope this helps, > ?John > > -------------------------------- > John Fox > Senator William McMaster > ?Professor of Social Statistics > Department of Sociology > McMaster University > Hamilton, Ontario, Canada > web: socserv.mcmaster.ca/jfox > > >> -----Original Message----- >> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] > On >> Behalf Of Shige Song >> Sent: November-13-10 10:14 AM >> To: r-help Help >> Subject: [R] Question about the "effects" package >> >> Dear All, >> >> I am using the "effects" package to produce predicted probability from >> a logistic regression. The graph looks really good. I soon realized >> that the y-axis is not spaced equally. For example, in my case, the >> distance between 0.02 and 0.04 is much greater than that between 0.06 >> and 0.08. I can guess a reason for this, but unfortunately, this >> distorts my story. Is there a way to change this and make the y-axis >> equally spaced? >> >> Many thanks. >> >> Best, >> Shige >> >> ______________________________________________ >> R-help at r-project.org mailing list >> 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. > > >