Assessment Tools: What They Do and Why You Should Be Harnessing Their Power

By David Ostberg, PhD

November 2008

Just as frontline call center agents are highly influential in forming lasting, positive impressions, high agent turnover rates significantly damage a call center's ability to maintain service levels and create a positive caller experience.  When the exorbitant costs of hiring, training, and lost productivity are added to the mix, it is easy to see why reducing agent turnover is a high priority.

In addition to agent turnover, the other dominant problem that hinders a call center's operation is suboptimal agent productivity.  In most cases, the root causes of both these challenges result from a poor job fit or a lack of the skills that are critical success factors for specific agent positions.  Selection assessments and screening tools can help address these issues by providing vital information about job applicants that directly relates to employee productivity and retention.  A call center's selection and hiring process should address the root causes of turnover by answering key questions such as: Are the agents a good fit for the job?  How well will they perform?  How long will they stay? 

In general, assessments are designed to capture information which relates to an individual's likelihood of behaving or performing in certain ways in the future.  These tools do so by measuring three different kinds of candidate data:

1) What can they do based on their personality, natural abilities, or aptitudes?

2) What have they done in previous jobs, activities, and educational or training settings?

3) What do they want to do in the future, based on their interests, motivation, and preferences?

What Applicants Can Do: As far as what applicants can do, assessments are designed to measure stable characteristics associated with a job applicant's personality and natural abilities.  For example, some jobs may be more suitable for individuals who are extroverted and less risk-averse, while others may require that an employee be highly detail-oriented and task-focused.  However, every human being is unique, with differences landing somewhere along a continuum of each of several personality and ability dimensions.  Assessment experts can develop tools to accurately measure where an individual falls on various job-relevant dimensions of personality or ability, creating valid predictions about how a person may behave in various situations.  The unique benefit to these types of assessments is that predictions can be made about future behavior in roles in which the applicant has no previous experience.  As a result, these "can do" measures are usually the most appropriate type of screening tool for hourly jobs and should comprise a significant part of the assessment solution. 

What Applicants Have Done: Concerning what applicants have done, assessments are designed to measure an applicant's previous experiences, past behavior, education or training, and accomplishments.  Examples include resume scoring for education, relevant work history and skills training, behaviorally oriented interview questions that focus on how an applicant handled a specific situation in the past, and job-specific knowledge or skills tests.  These types of assessments are based on the principle that future performance is best predicted by past performance.  For hourly jobs, the most appropriate "have done" tools include basic qualifications screening questions and behaviorally oriented interviews.

What Applicants Want To Do: Finally, regarding what applicants want to do, assessments are designed to measure differences in an applicant's motives, aspirations, preferences, and interests.  These types of tools are less effective for predicting productivity, but they are well suited to predicting job and culture fit, employee satisfaction, and retention.  Realistic job previews fall into this category, as candidates can take a "sneak peek" into the actual organization and job and opt out of the selection process if something looks particularly undesirable or ill-suited to their interests and expectations.  Measuring motivation and preferences is valuable for almost any level of job, but again, this primarily relates to employee satisfaction and retention rather than productivity and performance.

An Assessment of Assessments: Some common criticisms of selection assessments are that they don't work, they're not fair, and they're too "fake-able."  However, research indicates that a well-designed set of assessments can predict performance and early retention more accurately and fairly than traditional subjective approaches.  Assessments help remove any bias that hiring managers and recruiters unintentionally bring to the process.  Further, assessment item formats - such as those utilizing a balance of appealing or "socially desirable" response options - have been developed which truly minimize an applicant's ability to "game" the assessments or figure out which answer is ideal.  In fact, using closed-loop analytics, assessment developers can identify which items are susceptible to faking and can subsequently modify or replace them with content that is more effective.  Also, research demonstrates that low levels of faking in any part of the selection process (including face-to-face interviews) do not affect validity in a significant way.

Although a wide body of evidence demonstrates how powerful and accurate well-designed assessments can be, there are, unfortunately, many poorly crafted, inappropriately applied assessments on the market.  As a result, and in accordance with the Uniform Guidelines on Employee Selection Procedures and the Principles for the Validation and Use of Personnel Selection Procedures, organizations need to take the steps necessary to ensure that any assessment strategy they use is valid for their environment and jobs.  In addition to reviewing relevant job analysis reports and test validation manuals, test administrators need to confirm that the assessments are effective within their organization by analytically evaluating the impact the selection tools and processes have on their performance outcomes.

Further, because so many unproven, poorly designed assessments and selection "solutions" are available today, it is critical that organizations take the time to investigate how the tests were developed, and who developed them.  Although many vendors take a "one-size-fits-all" approach, no assessment is as effective and accurate as one that has been designed and validated by experts for specific positions within specific industries.  Because previous experience is less critical in most hourly jobs, a call center selection system should typically focus more on screening for behaviors ("can do") and motivation levels ("likes to do") and less on cognitive abilities and technical skills ("has done"), as might be the case with higher-level, salaried jobs. 

Conclusion: No single "perfect" solution exists for every selection challenge, but a well-thought out, well-designed process that uses selection science and incorporates multiple components - such as a realistic job preview, qualifications screening, validated assessments, and behaviorally oriented structured interviews - can dramatically improve the quality of hiring decisions.  As a result, organizations that choose to invest time and resources into evolving their selection processes will see a significant improvement in the productivity of their agent workforce, while improving retention and reducing the burden on their hiring and training staff.

Taking it one step further, call centers need to evaluate the impact that selection tools are having within their organization by harnessing the power of closed-loop analytics and on-demand reporting and should require that their selection solutions enable them to do so.  By improving call center agent selection and hiring, the right pieces can be put in place to drive measurable improvements throughout the organization and get ahead of competition. 

David Ostberg, Ph.D., is an Industrial and Organizational Psychologist with 10 years experience designing and evaluating selection systems for companies in service industries.  A member of the American Psychological Association and Society for Industrial & Organizational Psychology, Dr. Ostberg is currently the Vice President of Selection Science for Evolv On-Demand.