Employee Selection: References, Testing, and Performance Prediction

Overview of Employee Selection: References and Testing

• Instructor: Timothy R. Morales

• Core Objective: These notes explore the various methods used in personnel selection to predict job performance, focusing on references, academic background, ability tests, skill assessments, and personality/integrity testing.

Predicting Performance Using References and Letters of Recommendation

• Definitions:

  • Reference Check: This is the process of confirming the accuracy of information provided on a resume and job application.

  • Reference: The expression of an opinion regarding an applicant's ability, previous performance, work habits, character, or potential for future success. This can be delivered orally or through a written checklist.

  • Letter of Recommendation: A formal letter expressing an opinion regarding an applicant's ability, previous performance, work habits, character, or potential for future success.

• Reasons for Using References and Recommendations:

  • Confirming Details on a Resume: Used to verify the truthfulness of applicant information to combat resume fraud.

    • Supporting Evidence: A survey of UK employers (Reade, 2005) found that $25\%$ of employers had withdrawn job offers in the previous year due to lies on applications, and another $23\%$ had fired current employees after discovering resume fraud.

  • Checking for Discipline Problems: To determine if an applicant has a history of poor attendance, sexual harassment, or violence.

  • Avoiding Negligent Hiring: This refers to a legal situation where an employee with a previous criminal record commits a crime within the scope of their employment, and the employer is held liable for failing to check the employee's background.

  • Discovering New Information: References provide a deeper understanding of an applicant's personality and specific skills.

• Key Problems with References:

  • Leniency: Research (Aamodt & Williams, 2005) shows that letters are overwhelmingly positive. Less than $1\%$ of references rate applicants as "below average" or "poor" because applicants choose their own references.

  • Knowledge of the Applicant: Writers often do not know the applicant well or have not observed all aspects of their behavior.

    • Example: Professors may write letters for students they only know from one or two classes. More reliable recommendations come from instructors who have taught a student in multiple classes or interacted with them outside the classroom.

  • Reliability: There is often a lack of agreement between two people providing references for the same person.

    • Statistics: Research reveals reference reliability is only $0.22$ (Aamodt & Williams, 2005).

    • The Reliability Paradox: There is more agreement between recommendations written by the same person for two different applicants than between two different people writing recommendations for the same single applicant.

• Dynamics of Observed Behavior:

  • Behavior can be processed, remembered, and recalled differently by two different observers (Observer A and Observer B).

  • Categories of behavior include:

    • Behavior observed by both A and B.

    • Behavior observed only by A.

    • Behavior observed only by B.

    • Behavior observed by neither A nor B.

Predicting Performance Using Applicant Training and Education

• General Performance Trends:

  • A meta-analysis (Ng and Feldman, 2009) determined that better-educated employees:

    • Demonstrate higher performance.

    • Are more likely to engage in "organizational citizenship" behaviors.

    • Are less likely to be absent.

    • Are less likely to engage in on-the-job substance abuse.

• Police Performance Case Study (Aamodt, 2004):

  • Education was found to be a valid predictor of performance in the police academy ($r = 0.26, p = 0.38$).

  • Education was a valid predictor of on-the-job performance ($r = 0.17, p = 0.28$).

  • Education added incremental validity to cognitive ability tests.

Predicting Performance Using Applicant Ability

• Ability Tests: These tap into the extent to which an applicant can learn or perform a job-related skill. They are mostly used when applicants are not expected to know the job at the time of hire and will instead be taught necessary skills.

• Types of Ability:

  • Cognitive Ability: Includes dimensions such as oral/written comprehension, oral/written expression, numerical facility, originality, memorization, mathematical/deductive/inductive reasoning, and general learning.

    • Function: It predicts performance by allowing employees to learn job knowledge quickly and process information for better decision-making.

    • Tests: The Wonderlic Personnel Test is the most common industry test ($12$-minute administration, group-capable). Other examples include the Miller Analogies Test, Quick Test, Raven Progressive Matrices, and Situational Judgment Test.

  • Perceptual Ability: Measures facility with spatial relations and form perception. Includes vision (near, far, night, peripheral), color discrimination, depth perception, glare sensitivity, speech clarity/recognition, and hearing sensitivity/auditory attention/sound localization.

  • Psychomotor Ability: Measures facility with processes like finger/manual dexterity, control precision, multi-limb coordination, response control, reaction time, arm-hand steadiness, wrist-finger speed, and speed-of-limb movement.

    • Applications: Useful for surgeons, electricians, pilots, police officers, and assembly line workers.

  • Physical Ability: Used for jobs requiring strength and stamina (e.g., firefighters, lifeguards). Measured via job simulations or physical agility tests.

• The Nine Basic Physical Abilities:

  1. Dynamic strength: Strength requiring repetitions.

  2. Trunk strength: Stooping or bending over.

  3. Explosive strength: Jumping or throwing objects.

  4. Static strength: Strength not requiring repetitions.

  5. Dynamic flexibility: Speed of bending, stretching, or twisting.

  6. Extent flexibility: Degree of bending, stretching, or twisting.

  7. Gross body equilibrium: Balance.

  8. Gross body coordination: Coordination while the body is in motion.

  9. Stamina: Ability to exert effort over long periods of time.

Predicting Performance Using Applicant Skill

• Work Samples: Techniques measuring current job-related skills rather than potential ability. The applicant performs actual job tasks (e.g., a mechanic fixing a fan belt, a secretary typing a letter).

  • Advantages:

    • Excellent Content Validity (directly related to tasks).

    • Excellent Criterion Validity (predicts actual work performance).

    • Excellent Face Validity (applicants see the connection, leading to fewer legal challenges).

    • Lower racial differences in scores compared to cognitive ability tests.

• Assessment Centers: A selection technique using multiple assessment methods where multiple trained assessors observe applicants performing simulated tasks.

  • Common Activities:

    • In-Basket Technique: Simulates daily paperwork (memos, emails) found in a manager's "in" and "out" wire baskets. Applicants must prioritize and handle the tasks provided.

    • Simulations: Diverse activities like role-plays that place applicants in realistic job situations.

    • Leaderless Group Discussions: Applicants meet in small groups to solve a problem without an appointed leader. They are rated on leadership, cooperation, and analytical skills.

    • Business Games: Exercises demonstrating creativity and decision-making. (e.g., Developing a proposal for a new iPhone app including target audience, cost, and marketing).

Predicting Performance Using Personality, Interest, and Character

• Personality Inventories: Psychological tools measuring facets of an applicant's personality. They are divided into those measuring "normal" personality and those measuring "psychopathology."

• The Big Five (Five-Factor Model) Dimensions:

  • Neuroticism: Spectrum ranging from emotional lability/depressivity/shamefulness (High) to fearlessness/shamelessness (Low).

  • Extraversion: Spectrum ranging from excitement seeking/attention seeking (High) to social withdrawal/detached coldness (Low).

  • Openness: Spectrum ranging from magical thinking/eccentricity (High) to inflexible/close-minded (Low).

  • Agreeableness: Spectrum ranging from submissiveness/gullibility/selflessness (High) to manipulativeness/callousness/deceitfulness (Low).

  • Conscientiousness: Spectrum ranging from perfectionism/workaholism (High) to distractibility/irresponsibility/rashness (Low).

• Concerns: Because these are self-report inventories, they are relatively easy to fake.

• Tests of Psychopathology (Abnormal Behavior): Determine if individuals have serious psychological problems (e.g., schizophrenia, bipolar disorder).

  • Projective Tests: Provide unstructured tasks like describing inkblots or drawing pictures (e.g., Rorschach Inkblot Test, Thematic Apperception Test). These are rarely used in selection due to questionable reliability/validity and high cost.

  • Objective Tests: Structured tests with limited answers scored via standardized keys. The Minnesota Multiphasic Inventory (MMPI) is the most popular. Others include the Millon Clinical Multiaxial Inventory (MCMI) and the Personality Assessment Inventory (PAI).

• Interest Inventories: Designed to identify vocational interests. The Strong Interest Inventory uses $325$ items (like/dislike) to help in vocational counseling.

• Integrity Tests (Honesty Tests): Designed to predict an applicant's tendency to lie or steal. Commonly used in retail, fast food, and hospital settings.

• Conditional Reasoning Test: A method to reduce faking by asking test-takers to select the best justification for a statement. The choice made is thought to indicate the individual's aggressive biases or beliefs.