3 Lessons from The SIOP 2021 Conference

Every year the Society for Industrial-Organizational Psychology (SIOP) holds a conference featuring the most recent research, best practices, methodology, and much more in people research and people analytics. Every year that I go, I end up having more that I want to accomplish than I have time or energy to do. I attend many sessions on the use of artificial intelligence (AI) for various assessments, personnel selection processes, talent management, and use of organizational surveys. Thanks to everyone who organized these activities and prepared presentations on these important topics.

I hope that no matter what career stage or specific area of people analytics or HR, you’ll find something of use in the things I took away from the conference. Here is my list of 3 things I took from this year’s SIOP conference.

The Fundamentals Still Apply

There are many new things to learn and discover throughout the people analytics sphere, which gives us a lot to look forward to. But I sometimes worry that people forget the basic elements of good analytics, data stewardship, and responsible data management along the way.

One of the neat things I saw were presentations about using advanced AI methods to improve the traditional processes. I hadn’t even considered what those possibilities could be because the focus is always so driven on new methods. It doesn’t have to be something completely new in order to be impactful and meaningful for clients.

Then there was the discussion about succession management during the pandemic. It was very interesting to hear other consultants discuss companies who were woefully unprepared because they did not take the dearth of management talent seriously in their organization. When COVID-19 created havoc and people were leaving the workforce and reshuffling responsibilities, some companies had no idea who they would put into management positions and ended up making desperate hires instead of thoughtful ones. Not a good situation when the whole nature of work shifted.

It also brought to mind the research from one of my former professors, who found that when you match specific characteristics and specific jobs well, the results can double the impact on performance compared to broad matching schemes. If this sounds very basic, it should. The problem is that with so many shiny new ‘toys’ out there, we sometimes forget these principles and try to take shortcuts because the new things promise progress and advancement. The basics are still necessary no matter how far the field advances.

People Analytics is Best Done Collaboratively

The SIOP conference has had an increasing number of presentations on the use of new analytic methods such as artificial intelligence (AI). Although I will talk more specifically about my thoughts on AI methodology in the next section, there is a broader point to this topic that I came away with. That is the point of collaboration. I think that too often, professionals working in the people data space are missing chances to work together on issues. Data scientists are advancing and creating new ways to look at organizational data. However, there is a valuable perspective in using that data for valid prediction and (in some cases) using it responsibly that I-O psychs can add to the process.

Several conference speakers brought up the fact that the use of AI in things like hiring, promotions, and career development opportunities are going to happen regardless of the involvement of I-O psychs. Sometimes the data scientists don’t have the context and knowledge of working within organizational decision making like I-Os do. And sometimes, I-Os don’t have enough of the specific technical skill to work on these algorithms. And then there are a whole host of other stakeholders that add valuable input to the process that we forget about. Combining these skills sets and using the best of what everyone has to offer would go a long way towards getting buy in from organizational leaders, reducing user anxiety about new technology, and easing fears of policy makers that this will adversely impact applicants. That collaboration has to take place across the academic/practitioner as well as the data science/I-O cohorts. Ultimately, I left the conference feeling like there is a lot more in common between the analysis methods than I thought there were. That should provide a sufficient foundation on which the various groups can build.

Be Thoughtful About Artificial Intelligence

The world of AI provides so many interesting avenues and applications for people analytics. I attended several sessions on the topic to try become more aware of what the capabilities and what the concerns might be. On the capabilities side, there is quite a lot. There was research outlining the possibility of using AI to create personality assessments, reduce bias in assessments using facial recognition, better predict performance, understand career development paths, and analyze interview data more efficiently and accurately. There is a growing body of evidence to suggest that these are viable solutions.

However, it was also abundantly clear to me that there is still so much more needed to establish widespread validity of these methods and do it in a way that is comfortable for applicants and decision makers . New laws are beginning to restrict the ways in which these methods can be applied. There are still concerns with understanding the underlying calculations and how those algorithmic decisions are made in the case that algorithms result in biased decision making. Additionally, there are concerns about the ways in which some data science organizations are attempting to apply the algorithms (see above). In some cases, AI may not be getting any more information than the currently available methods. I also wrote about this topic here: Algorithmic Justice.

So, the overall message I took from this was to be very purposeful and thoughtful about the goals you have for your work and what these new tools can do to assist in those efforts. If you’re using AI just because it’s the new thing, you will be disappointed. There are ways that AI can currently be used, but understand there are still a lot of limits and boundaries for using it appropriately. This should provide a reasonable guiding principle for any people analytics endeavor. I heard this same messaging in sessions regarding balancing science and practice, succession management, employee surveys, and career development.

Final Thoughts

It’s been a couple years since I have attended SIOP. Just before SIOP 2019, I ruptured my Achilles tendon and was having surgery at the time of SIOP. Last year, I opted out of SIOP because the pandemic created so much uncertainty that I needed to focus in on everything else going on in my life. Now, 3 years since my last SIOP conference, I realized how much it can provide. Not only did I catch up with old friends, but I made new connections as well.

There is a sense of community in the I-O psych realm that I think is very difficult to replicate outside of this conference. It is inspiring to see others doing great work. It is comforting to know that others have struggled through similar issues to the ones I face as a consultant. It is always good to see old friends. This is a network that I think I need more than I thought was necessary.

Above all the SIOP conference constantly reminds me that professional development is the career arc that one takes, not just a few moments of training. It is the continual process of finding growth and fulfillment through the various ups and downs of work. Several years out of my PhD, I’m still finding all kinds of things I need to learn and extra things that I would like to try. And although I am continually seeing research skills from the next generation of I-O psychs that suggests that they could take my job if they wanted it, I find myself more than ever wanting to provide them the opportunities to prove that they can do so. Because someday they will take my place whether any of us want that or not.

As always, thanks for reading!

–Brandon
Subscribe to the blog below!