Welcome to the home for my symposium presentation: *Is it time for the AAQ-II to move over?* I gave the talk at the 2019 conference of the LaMiss Chapter of the Association of Contextual Behavior Science. The webpage for the conference lives [here]( https://lamissacbs.org/). In this OSF project, you’ll find a copy of the PowerPoint file I used during the talk, as well as supplemental materials concerning the data analysis. As you’ll see, I’m not one for adding notes to my PPT slides. Here's the **primary message of the presentation**: At least as far back as 2011 (i.e., the year the AAQ-II was published in peer review, see [here](https://www.sciencedirect.com/science/article/pii/S0005789411000888)), scholars have criticized the AAQ-II. In one of the most recent criticisms, [Tyndall and colleagues (2019)](https://www.sciencedirect.com/science/article/pii/S2212144718302618) found the AAQ-II correlated more strongly with measures of anxiety, depression, and stress than it did with a measure of experiential avoidance. The findings suggest a major problem with convergent and divergent validity. The measure of experiential avoidance was the Brief Experiential Avoidance Questionnaire (BEAQ; [Gamez et al., 2014](https://psycnet.apa.org/record/2013-33700-001)). At the time of this writing, I’m a postdoctoral fellow at the [VISN 17 Center of Excellence Research on Returning War Veterans](https://www.mirecc.va.gov/visn17/), which specializes in the study of the development, progression, and treatment of PTSD in US veterans. My PI, [Eric C. Meyer](https://scholar.google.com/citations?hl=en&user=i99DmhQAAAAJ), heads an ongoing panel study on quality of life, functioning, psychiatric symptoms, and other constructs in this population. Eric is an ACT researcher and has also been tracking psychological inflexibility (e.g., with the AAQ-II) and experiential avoidance (e.g., with the BEAQ). The data from this particular study are from 501 post-911 war veterans. Within each veteran, data were collected at four time points over a two-year period (i.e., once every 8 months). Data collection started in 2014 and concluded in 2018, with start times varying across participants. The original goal of my presentation at the symposium was to do something of a conceptual replication of Tyndall’s analysis. We have the AAQ-II at two time points (i.e., baseline and 2-year follow-up) and the BEAQ at all four time points. In addition, we had measures of PTSD (i.e., [PCL-5](https://onlinelibrary.wiley.com/doi/abs/10.1002/jts.22059?casa_token=l4L-tmJdTRIAAAAA:ZV05eJE5ckQf_1YZxVly4FYLpfcf09KBM_Kq8mFSm2f8dYcZFsbAsthd2A1Y4E8bX30tMqkIeyh0npU)), depression (i.e., [PHQ-9](https://onlinelibrary.wiley.com/doi/abs/10.1046/j.1525-1497.2001.016009606.x)), functioning (i.e., [WHODAS-2.0](https://www.scielosp.org/scielo.php?pid=S0042-96862010001100010&script=sci_abstract&tlng=fr)) and many more. However, the direction of the study took a turn once I began the preliminary analyses. Before the primary validity analyses intended to augment and extend those from Tyndall and colleagues, I performed a series of factor analyses to ensure the psychometric integrity of the measures. The AAQ-II performed as expected; it showed reasonable fit with a single-factor solution, particularly when including a few of the residual correlations seen in the prior literature. Unexpectedly, though, the BEAQ showed rather poor fit with the theorized single-factor structure. The BEAQ is a brief version of the earlier Multidimensional Experiential Avoidance Questionnaire (MEAQ; [Gamez et al., 2011](https://psycnet.apa.org/record/2011-08823-001)), which showed a 6-factor structure. Based on the original structure, I compared the 1-factor solution with a * simple correlated 6-factors solution, * a hierarchical extension of the same, and * an alternative bifactor model with one general factor and 6 residual factors. For each of the 4 time points, the simple 6-factor structure solution was the clear winner. In an exploratory spirit, I also considered a couple models for eight of the 15 BEAQ items. However, based on the poor performance of the BEAQ for providing a unitary measure of experiential avoidance, my final recommendation was for experiential avoidance scholars to back up and reassess whether the 15 proposed items for the BEAQ are the best ones to use for a unitary measure. I suspect they are not. Researchers are free to use our 8-item solution as a short-term stop gap. However, I do not believe this is a reasonable long-term solution. Anyway, the unexpected performance of the BEAQ derailed any assessment of the relative performance of the AAQ-II in relation to the other measures. My conclusion is that **though the AAQ-II has shortcomings, the current version of the BEAQ is not a reasonable replacement. If researchers are adamant the AAQ-II is deficient, they will need to look elsewhere for a suitable successor**.