Incorrect in the presupposition that what is scientifically measureable, or what is chosen to be measured, equates directly to an individual's personal outcome, or that a predetermined outcome is always more desireable than what actually emerges.  Scientific research has as *its* outcome, a statistical consistency, no better, no worse.  But if something much better but not categorically equivalent to the outcome arose, would the research even be able to recognize it? And when we do scientific research, in terms of behavioral sciences, what is it that we actually mean?  Isn't it really the realm of control freaks and insurance providers? Communication efficacy is hard to pin down.  How does one, for instance, estimate how effectively Picasso communicated in Guernica?  Or Madonna, singing "Like a Virgin" while she rolled across the stage in a wedding gown several years ago at the Grammies?  Who's to say that not more than one sufferer of impotence got a rise out of that the first time in ages?  But try to repeat that in a statistical setting and you'll drain all the blood out of it, if you know what I mean. btw when I said that the meta-model covers a lot of what rebt covers, I was simply talking about what one might possibly uncover while attempting to utilize either of the models in actual practice, apart from whatever theoretical differences there might be in the design of each model.