I can't speak for Pragma, but for me "basic competency" is less about running the analyses and more about being able to read (and understand!) analyses. If someone is planning on going 100% clinical - that is what I think the goal should be, and I think that was the original intent of the Vail model (though many schools have obviously pushed that to giving doctorates to people with undergrad-level knowledge).
t-tests, mean/median/mode are the bare, bare basics - I'd worry if someone got ADMITTED to grad school without knowing that, its shameful when UGs don't know it. ANOVA/Regression I would also place in the "basics" category, as most other techniques are based on these. I'd go a little further than that before declaring someone has "basic competency" in stats. Some things (post-hocs, power analysis, etc.) are important to understand. This is a personal belief, but I think especially for folks planning on clinical careers, meta-analysis is an absolute necessary. Again, I'm not talking about knowing how to compute power for a meta-analysis of imaging data with nothing but a calculator, but given its role in EBP I think a basic foundation (knowledge of common effect size measures, how to catch common flaws, etc.) is crucial.
Basically - I'd say its sufficient when one can read the bulk of the literature in top tier journals and "get" the results section, at least noting any major flaws (i.e. why are they running ANOVA when they have a categorical DV?), even if some of the more nuanced issues elude them.
After reading though this thread, I think the discussion on dissertations, while warranted and interesting, misses the mark. However, I think Ollie captures the importance of the issue at hand.
From a professional development standpoint, the completion of dissertation can serve a variety of purposes. We need to consider long-term career goals.
For someone interested in a career in academia, a thesis and/or dissertation can serve as a springboard to posters, oral presentations, and publications. One's level of productivity, as well as their ingenuity in applying experimental methods and the originality of their research program, are essential aspects of building an academic career. Consequently, I would suspect that individuals with these particular goals would pursue a much more extensive and nuanced understanding of clinical science. Greater effort in studying experimental methodology, basic and complex statistics, and basic psychological science would all be critical components. The nature, depth, and complexity of their dissertations should reflect their goals. Would everyone agree?
For one interested in a career in clinical practice, a thesis and/or dissertation may serve an entirely different goals. To be specific, the development of skills that allow one to identify relevant research to their scope of practice, comprehend the findings of such work, and critically evaluate the findings and methodology can be seen as much more desirable than publications. From this point of view, the dissertation provides an opportunity critically evaluate the literature, study and apply experimental methodology, implement statistical analyses, interpret data, and highlight how the results fit into the broader context of our field. In this light, the dissertation promotes the acquisition of knowledge in these areas through guided experiential learning, whereby a student applies basic skills in an effort to understand the research process. Within the Vail framework, this truly helps one develop competency in the consumption of research.
To Ollies point, the ability to comprehend the essential aspects of treatment outcome research and related clinical research is what is important. Having said that, I think we can all agree that comprehending complex statistics (i.e., SEM, latent class analyses) is far less important to the practitioner than their ability to comprehend more basic statistics (i.e., t-tests, ANOVAS, effect-sizes, odds-rations, survival analyses) that are used frequently in treatment outcome studies. Yet, for the academic, SEM is becoming a more frequently used tool that allows for the evaluation of theories and hypothesis. Are the bulk of studies practitioners read going to employ such complex statistics? Probably not. Is the practitioners ability to understand such statistics influence their ability to practice? Probably not. Is there ability to understand experimental methodology going to influence there ability to formulate a empirically based treatment plan? Absolutely.
We need to consider the contexts in which people work and the knowledge they need to acquire in order operate at a competent level. So while qualitative and quantitative differences in dissertations exist across PhD and PsyD programs, we need to consider the goals of the students and how their education prepares them for those goals.
