HEALTHCARE HIRING PERSPECTIVES BLOG

Attributes that Predict Nursing Success: A 16,000 Nurse Validation Study

Posted by  Bryan Warren

predict-nursing-success.jpgSeveral years ago, I met with a nursing manager about her hospital’s process for hiring nurses. Hospital leadership had an idea that they could do a better job hiring nurses. Not “attracting” nursing candidates. Everyone is always trying to improve their ability to attract qualified candidates.

Hospital leaders and HR leadership felt, rightfully, that they could take a more deliberate, evidence-based approach to selecting candidates who were more likely to succeed. Data indicated their feeling was on point. They were not particularly adept at choosing the candidate who would succeed.

This nursing manager had a simple solution. Let HER do every interview. “I don’t need tools or an interview guide. I need a resume and a five-minute interview to determine whether someone is a good nurse or not.” I was impressed. If this were true, we were on to something. Talk about a quick, efficient, cost-effective way to build a top-notch team!!! We should pull this nursing manager from whatever else she is doing and have her just do interviews all day – simple.

“I don’t need tools or an interview guide. I need a resume and a five-minute interview to determine whether someone is a good nurse or not.”

Unfortunately, it didn’t take long to figure out that her faith in her abilities was not supported by the data. She had as many bad hires as anyone in the building and turnover in her department was actually higher than the rest of the hospital! This should tell you something about the unfounded faith managers have in their ability to read people.

Related: Traditional Job Interviews Are Useless

In recent years, more healthcare talent professionals are realizing something has to change. In an era of big data, we need to be more objective, more consistent, and more effective at predicting candidate success.

I’ve seen a lot of results published by vendors in our space. They report that their resume review tool, their interviewing tool, or their pre-employment test predict success. They identify a hospital that uses their tool and has high patient satisfaction or even patient quality results and then conclude that their tool works.

I didn’t pay a lot of attention in my statistics class but I recall that correlation does not equal causation. If they look at a larger number of healthcare clients and did a real validation study of some sort, they MIGHT show a real, causative correlation between use of certain strategies and tools and specific results, but that’s most definitely NOT what they are doing. What they are doing is not scientific by any means – it’s quite disingenuous, to be honest. What they aren’t CAPABLE of doing is a real study of data to show their value so they fall back on this trick.

Related: Are You Hiring a Nurse or a Pirate? Face Validity in Healthcare Assessments

The most DIRECT and honest way to evaluate efficacy of individual selection tools is to look at individual performance. Our Research and Development team does this routinely, and our most recent study revealed some interesting information:

  1. Certain key individual attributes are most predictive of on-the-job nursing performance; and

  2. Those that are somewhat less predictive reveal a cultural gap.

16,000 Nurse Validation Study

In this study, we looked at 16,000 nurses hired over a one-year period. We had their scores from the hiring process, specifically, their scores on a healthcare-specific hiring tool measuring key behavioral skills and values. We also had supervisor ratings of on-the-job performance. Through statistical analysis, we can understand meaningful correlations. For instance:

  • Candidates who scored high on the test in the area of Excellence and Innovation were three times more likely to be rated as a top performer.

  • Alternatively, those who scored low in this attribute were four times more likely to turn over in their first year.

What does this mean? This organization values innovation, adaptability, and those who strive for excellence. If you score high in this area, you are far more likely to be successful. If you score low, you will struggle. The same is true for Responsibility and Integrity:

  • Nursing candidates who scored high in Responsibility and Integrity were 2.7 times more likely to be rated as a top performer.

  • If they scored low in this area, they were 2.5 times more likely to turn over in year one.

A surprise finding? Scoring high in Caring and Listening (which includes empathy) was only 1.8 times more likely to result in a high on-the-job rating. In other positions, it was actually neutral – i.e. had no impact on predicting success.

This is not a complete surprise to us. We’ve seen this before. Depending on culture – what managers value – high levels of empathy may or may not result in high performance ratings. It means that success in taking care of patients is complex and empathy/caring is not enough for success. It might also mean that the organization’s culture still has some room to grow and manager training may be in order.

Related: Culture Fit is Determined by the Behaviors the Organization Values Most

THESE are evidence-based hiring results. THIS is what it means to understand data as it pertains to identifying candidates with the greatest chance of success. THIS is the sort of data analytics organizations need to be doing to make the most of their talent strategies.

To Learn More – see our free white paper on Evidence-Based Hiring:

healthcare hospital behavioral competency model

Tags:   nurse hiring strategies, evidence-based hiring, healthcare hiring, turnover, hiring process, nursing turnover

Bryan Warren

Bryan is the former Director of Healthcare Solutions at Select International. He was responsible for developing and promoting tools and services designed specifically for the unique challenges faced by healthcare organizations.

Subscribe to Email Updates

Recent Posts

Four Hospitals Save $1.3 Million by Cutting Service Worker Turnover in Half