Academic journal article Public Personnel Management

Predicting Performance with Letters of Recommendation

Academic journal article Public Personnel Management

Predicting Performance with Letters of Recommendation

Article excerpt

In psychology, it is commonly believed that the best predictor of future performance is to look at past performance. Thus, if an organization wanted to hire a fire fighter, the best applicant might be one who has not only previously been a fire fighter, but one who was also a successful fire fighter in his/her previous job.

While it is not very difficult to verify the previous employment of an applicant, it can be rather difficult to verify the quality of his/her previous performance. The authors of this article recently watched the National Football League (NFL) draft on television and were envious of the fact that professional football teams could assess a college player's previous performance by watching game films. That is, the football scores did not have to rely on opinions of the other coaches. Instead, they could watch every minute that the player had played while in college.

Unfortunately, very few applicants bring "game films" of their previous employment performance. Instead, an employer must obtain information about the quality of an applicant's previous performance by asking an applicant either to supply names of references that the employer can call or to provide letters of recommendation from previous employers.

Even though references are commonly used to screen and select employees, they have not been successful in predicting future employee success (Muchinsky, 1979). In fact, the average validity coefficient for references is only .13 (Browning, 1968; Mosel & Goheen, 1959). This low validity is due mostly to four main problems found with references and letters of recommendation: Leniency, knowledge of the applicant, low reliability, and extraneous factors involved in the writing and reading of letters of recommendation.


Choice of References - Research is very clear on the fact that most letters of recommendation are positive (Carroll & Nash, 1972; Yoder, 1962; Myers & Errett, 1959). Because at some point we have all worked with terrible employees, it would at first seem surprising that references are so positive. However, it should not be surprising when it is kept in mind that applicants choose their own references. If given the chance to choose their own references, even undesirables such as Nazi leader Adolph Hitler, serial killer Ted Bundy, and terrorist Abu Nidal would be able to find three people who would provide them with favorable references.

Confidentiality of the Reference - A second factor that will increase the leniency of references is the confidentiality of the reference. By law, people have a right to see their reference letters, but by signing a waiver, they can give-up that right. Research by Ceci and Peters (1984) and by Shaffer and Tomarelli (1981) has indicated that references tend to be less lenient when the applicant waives his/her right to see the letter of reference. That is, if a person writing a reference letter knows that the applicant will be allowed to see the letter, the letter writer will be more inclined to say nicer things.

Sex and Race - Two other minor factors which affect the leniency of references are the sex of the letter writer and the race of the letter reader. Carroll and Nash (1972) found that female letter writers are more lenient when referring female applicants while Bryan (1989) found that black professionals are more lenient in evaluating the contents of letters than are white professionals. The combination of applicants choosing their own references, retaining the right to see a reference letter, being referred by a female reference, and being evaluated by a black professional will make the reference letter much more positive than it should be based on the applicant's actual performance.

Knowledge of the Applicant

A second problem with letters of recommendation is that often, the person writing the letter either does not know the applicant well or has not observed all aspects of the applicant's behavior. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.