While it may be tempting to hire the freshest face with the glowing interview, the best choice, time and again, is the above-average employee in the organization who has consistently been rated well in the past, according to a new study by a Cornell University researcher and his colleagues. The researchers provide estimates of the strength of the relationship between past and future performance that supervisors can use in the hiring process.

�Past-performance appraisals are by far much more valid in predicting future job performance. Thus, as in most cases where there is reliable, valid information about an above-average internal candidate, that candidate should be selected over top external candidates who might have had a stellar interview,� says Michael Sturman, an associate professor of organizational management, communications and law in Cornell's School of Hotel Administration. Past-performance records, he says, are much better predictors of future performance than other recommended, but less valid, selection tools, such as structured interviews, cognitive-activity tests and job simulations. This is especially true, Sturman notes, for professional positions such as managers and supervisors, but it also holds true for jobs such as sales positions, bank tellers and production workers.

Sturman's report, �How to Compare Apples to Oranges: Balancing Internal Candidates' Job-Performance Data with External Candidates' Selection-Test Results,� co-authored with Robin A. Cheramie and Luke H. Cashen, doctoral candidates at the E.J. Ourso College of Business Administration at Louisiana State University, was published by the Center for Hospitality Research at Cornell's Hotel School in September. The full report is available free at www.hotelschool.cornell.edu/chr/research. The report also was published in the August 2002 issue of the Cornell Hotel and Restaurant Administration Quarterly.

Even though an employee's job performance may vary over time as the worker acquires experience, gains or loses motivation and has different opportunities to succeed or fail, employers can expect an employee who has done well in the past to deliver a solid, positive performance record for years to come, the researchers found.

Sturman and his colleagues analyzed 20 human-resources studies on individual performance over three or more time periods published in a range of scholarly journals on management, marketing and psychology. Together the studies' findings comprised hundreds of observations about on-the-job behavior in a wide range of positions. The researchers then calculated the correlation between past appraisals and current performance in several different job types, including both professional and hourly positions in a range of fields, such as academia, sales, banking, insurance, manufacturing and finance. The range of jobs included both those that are evaluated objectively and those that are evaluated subjectively.

The �Apples and Oranges� report offers mathematical approaches employers can use to calculate the weight that should be accorded a particular type or age of appraisal. For example, Sturman and his colleagues found that the older the appraisal of an employee, the less it should be weighted, but on average, the validity of past employee data, even relatively old data, tends to be much more accurate than selection tests or interviews.

�Hiring decisions should be made based on whatever tool provides the most accurate prediction of future performance,� advises Sturman. �To do that, the information should be weighted by its validity.�

The researchers also caution employers to use statistical or other hard evidence rather than emotions or �gut feelings� when making hiring decisions. For both internal and external candidates, hiring data should be collected in a consistent and even-handed fashion, says Sturman.

The Cornell Hotel and Restaurant Administration Quarterly, of which Sturman serves as an editor, is considered the premier journal of applied research serving the hospitality industry and is published under the aegis of Cornell's Center for Hospitality Research. The center conducts and sponsors research studies aimed at improving the hospitality industry's fundamental operating knowledge.

Visit this link.