October 2016


Volume 54     Number 2    Fall 2016      Editor: Tara Behrend

Meredith Turner
/ Categories: 542

Using Science to Choose the President

Jessica Deselms, Lauren Bahls, Kristie Campana, and Daniel Sachau

What if the president of the United States of America was selected rather than elected? In other words, what if candidates applied for the job of president in much the same way that executives apply for the job of CEO? What would the selection process look like? What knowledge, skills, and abilities would the ideal presidential candidate possess? I-O psychologists are all over the place in government work. For example, SIOP’s GREAT committee advocates for evidence-based decision making in government. Many I-O psychologists work within government agencies such as the TSA, the FBI, OPM, and the Social Behavioral Sciences Team. There are also plenty of I-O organizations that often subcontract with the government, including FMP, HumRRO, and PDRI. Clearly, I-O psychologists are deeply involved in many government processes. Why not selecting a president? We posed these questions during interviews with a variety of SIOP members.  We are not suggesting a change in the Constitution.  Consider this as a kind of thought experiment in employee selection.


Job Analysis


All of our respondents thought that, as in any selection process, job analysis should be the first step. Let’s find out what a president actually does. Here is where the problems start. First, there is no official job description for the president, and the Constitution only offers 322 words on the matter. Next, only a very small number of people have access to the day-to-day activities of a president. Further, the public’s understanding of the president’s job is clouded by political pundits who opportunistically inflate or minimize the president’s role in world events. Does a president have control over our economy? The answer depends on which party you ask and how well the economy is doing. It is safe to say the president’s duties are varied. Presidential diaries tell us that a typical day in the life of president might include signing legislation, receiving security briefings, attending an awards ceremony, hosting dinner parties, and meeting with heads of state.


What Constitutes Successful Presidential Performance?


Even if we could get our hands on a comprehensive job description, a criterion problem looms large. What criterion should be used to measure the success of the president? As Doug Molitor (3M) states, “One person’s good president is not necessarily another person’s good president.”  Rob Ployhart (University of South Carolina) echoed this concern, “What we want the president to do is divided by different stakeholders.” Setting benchmarks for presidential performance is further complicated by the tenuous link between policy and effect. Success and failure are not immediate. For instance, Kasey Guentert (Korn Ferry Hay Group) notes that people still debate the success of FDR’s strategies, and there may be “economic repercussions from his policies that have yet to happen.” Another problem with performance criteria is that they shift.  Eric Heggestad (University of North Carolina Charlotte) noted that in 2008 the country was seeking a candidate who could lead us out of a financial crisis. It appears that in 2016, the key criterion is a candidate who can navigate divisive social issues. A third problem with the criteria was noted by Nancy Tippins (CEB), who stated, “The job shapes the president, and the president shapes the job.” Job skills play out differently for different presidents. Charm and warmth worked well for Reagan but not so well for Carter. Lyndon Johnson, generally unlikable, managed to get just about anything he wanted through Congress. So, similar skills deployed differently can lead to vastly different outcomes. Finally, another problem is that the measure of success is determined by voters.  As Molitor put it, what makes a good president is whether he or she is “able to make people think he or she is a good president.” Jose Cortina (George Mason University) was more direct: An effective president might be one who is good at “snowing voters.”


Let’s say we could overcome the criterion problem and we decide to conduct a job analysis, there is yet another roadblock: Who should do it? We know that who we choose as SMEs matters a great deal (Morgeson & Campion, 1997). The choice of SMEs is not obvious in this case. Dick Olson (Olson Consulting) and Carol Lynn Courtney (Courtney Consulting) noted that former presidents and selected members of Congress likely have intimate knowledge of the president’s job (for the curious reader, Obama recently outlined his typical day on The Oprah Winfrey Show, which you can view here: https://www.youtube.com/watch?v=zBlToJwaD4g). Leaetta Hough (HirePayoff, Dunnette Group) suggested polling a diverse group of constituents. Doug Molitor recommended that the chief of staff, Joint Chiefs of Staff, or the press secretary would be excellent resources for job information. We will not pretend that party lines wouldn’t be a problem.


Presidential Tests


Even though it is difficult to define good presidential performance, and Republicans and Democrats would spend years arguing over who should do the job analysis, our colleagues had a number of interesting recommendations for tools that could be included in a presidential selection battery. Biodata, job experience, and archival data were popular suggestions. Among others, Paul Sackett (University of Minnesota) and Tippins cited the adage that “past behavior predicts future behavior.” Courtney identified some key differentiators, such as international experience and cultural competency. Heggestad proposed coding public records for key behaviors.

            A number of our respondents suggested, quite reasonably, that presidential candidates should be intelligent, but even this came with caveats. Ployhart, Sackett, and Chuck Lance (Organizational Research & Development) warned about issues with range restriction. Despite what Huffington Post or Fox News grumpily insist, most presidential candidates are bright.  A number of people noted that in addition to focusing on raw intelligence, it would be useful to examine cognitive processes.  For instance, Cortina mentioned that it would be wise to examine candidates’ epistemology: Does the candidate believe something to be true because of evidence or based on wishful thinking?  

            Personality assessments were popular recommendations. Most respondents believed that the Big Five might be too broad, but Guentert suggested that the ambition and drive subfactors could be helpful. Rubenzer, Faschinbbauer, and Ones (2000) conducted a clever study where they asked presidential biographers to complete the Big Five for their respective presidents.  They found that presidents were rated lower on agreeableness, lower on openness to experience, and higher on extroversion than the typical person. 

            SIOP members’ suggestions for presidential skills and abilities also included tolerance for ambiguity, decisiveness, negotiation skills, and social skills. Most of our colleagues identified some component of integrity, commitment to country, and ethics. Olson and Courtney suggested that compassion, self-understanding, and “not taking yourself too seriously” might be keys to presidential success. Heggestad suggested that it would be nice to see the MMPI profile of candidates but recognized the legal pitfalls of that suggestion. Regarding the suggestion of integrity, we would certainly prefer our presidents not take bribes, but what other behaviors are relevant? Does integrity in one’s personal or family life apply? Lance wryly questioned whether we would want integrity at all; presidential business can be dirty work, and someone who is shrewd and calculating may ultimately make the decisions that would be in the United States’ best interest.

Presidential assessment centers were a common suggestion. This makes good sense, considering that many CEOs go through these types of assessments, and they offer incremental prediction over cognitive tests (Krause, Kersting, Heggestad, & Thornton, 2006). Given that the president is much like the CEO of the government, this approach might be a sensible fit. Sackett referenced former SIOP President Ann Howard, saying, “if it exists, it can be measured; if it can be measured, it can be measured with an assessment center.”  There was great overlap regarding the types of exercises that could be incorporated into an assessment center and competencies to examine, including negotiation, interface with foreign dignitaries, natural disaster, and handling of day-to-day activities. Certainly developing a presidential simulation would be complex, and would likely need to encompass both mundane and critical situations. As we have seen during Obama’s presidency, achieving cooperation among diverse stakeholders is of critical importance. A role-play exercise would need to illustrate how candidates can build consensus with rival political groups, negotiate with foreign leaders, and work with the press secretary to make decisions about how to keep the public appropriately informed. Creating a scenario for this with appropriate complexity would be difficult. Tippins and Molitor mused as to what a “Bay of Pigs” simulation would look like, and how the current candidates would handle it.  

As an interesting aside, it appears that presidential simulations have been created in the past. Typically, they are intended as a way to ease the transition from one president to the next. In 2012, former Governor Mike Leavitt designed a “Romney Readiness” team that was intended to be a miniature government that Romney could explore should he win the election. Clearly, they never had a chance to launch this initiative, but Minnesota Public Radio provided a brief interview with Leavitt, which you can find here: http://www.npr.org/2016/08/01/ 488274034/white-house-begins-transition-planning-4-months-ahead-of-election-day. This was meant more as a training exercise, but it would be interesting to translate this concept into a selection simulation.

One key consideration that came up again and again was the issue of leadership. A number of interviewees argued that the president has limited power and may have actions blocked by Congress or the Supreme Court. Along with leadership neutralizers, many of our colleagues discussed the importance of leadership substitutions. Sackett noted that the role of president is so immense, that nearly everything must be delegated to the cabinet or other staff.

            Heggestad and Hough specifically referenced transformational leadership. Hough noted that key components of the theory, such as inspiring trust, innovative thinking, maintaining and building relationships with diverse people, and visionary thinking would all be helpful in the presidency. Researchers have suggested that President Obama, on the campaign trail, used charisma and inspiration to gain the presidency with relatively little experience; whether he continued to be transformational in the White House is a subject of great debate (Bligh & Kohles, 2009).  On a related note, a number of respondents mentioned that the KSAs it takes to win a nomination might be quite different from the KSAs needed to be an effective president.


The Tradition of Elections


It is worth noting that a few of our interviewees insisted that presidential selection would be ill advised and that the election should stay in place.  For instance, Ben Schneider (Schneider Consulting) is an advocate for the current system.  He stated that the "trial by fire in the political arena” approach is the one to continue. Lance also supports the election system, arguing that it is a “moot point—presidents serve at the whim of the electorate.” Naturally, supporting the validity of our system will be difficult, with our small N (we would hopefully only add one data point every 4-8 years). Perhaps synthetic validity could be pursued, but linking CEO performance to presidential performance will be a tricky task and may miss key components of the president’s job that are not echoed in the private sector.

 Many other respondents indicated that they find comfort in an election. As Ployhart stated, “At the end of the day, if it was truly for the president, I would still want an election… If in our hypothetical world we gave [our selection battery] to every presidential-eligible citizen in the country and picked the person who comes out the highest… would you be comfortable without having an election?” Alas, these colleagues will ultimately have their way. As we near the end of this election cycle, we will all cross our fingers and hope that the successfully elected candidate knows what he or she is doing.  Let’s hope that we don’t find ourselves longing for a presidential selection!




Bligh, M. C., & Kohles, J. C. (2009). The enduring allure of charisma: How Barack Obama won the historic 2008 presidential election. Leadership Quarterly, 20, 483-492.

Krause, D. E., Kersting, M., Heggestad, E. D., & Thornton, George C., III. (2006). Incremental validity of assessment center ratings over cognitive ability tests: A study at the executive management level. International Journal of Selection and Assessment, 14(4), 360. doi:http://dx.doi.org/10.1111/j.1468-2389.2006.00357.x

Morgeson, F. P., & Campion, M. A. (1997). Social and cognitive sources of potential inaccuracy in job analysis. Journal of Applied Psychology, 82, 627-655.

Rubenzer, S. J., Faschingbauer, T. R., & Ones, D. S. (2000). Assessing the U.S. presidents using the Revised NEO Personality Inventory. Assessment, 7(4), 403-419. doi: 10.1177/107319110000700408


More Reading


Rubenzer, S .J., & Faschingbauer, T. R. (2004). Personality, character & leadership in the White House: Psychologists assess the presidents. Washington, DC: Brassey's.

Simonton, D. K. (2002). Intelligence and presidential greatness: Equation replication using updated IQ estimates. In S. P. Shohov (Ed.), Advances in psychology research (pp. 143-153). Hauppauge, NY: Nova Science Publishers.

Previous Article SIOP in Washington: Advocating for I-O in Federal Public Policy
Next Article Toward a Critical I-O Psychology
1366 Rate this article:
No rating