Section Fouranalysis are evolving to deal with situations in which there is
extensive teamwork or flexible work assignments.^37
Interviews
Interviews are a universal feature of selection systems. Recent
research has shown that interviews can have substantial
validity, particularly when they have more structure and are
conducted appropriately. One very large-scale research study
found a validity coefficient of .47 for interviews and .62 when
more structured interviewed were used. (Validity coefficients
range from 0 for no predictive ability to 1.0 for perfect
predictive ability.) Thus, the predictive ability of interviews can
be increased with more structure, which means that the
interviewers ask the same questions of applicants so that they
can be compared on the same dimensions. Other suggestions
for improved inter-viewing include restricting the range of
dimensions to be assessed to a small number of skills and
abilities and knowledge that is most important for the job.
Another suggestion is to limit the amount of information about
the applicant that is reviewed by the interviewer before the
interview. By limiting this information, the interview remains an
independent predictor and the interviewer is less biased by
information obtained from other means. In addition, the use of
multiple interviewers helps improve predictive validity, although