Employee Selection – Techniques

Why Employee Selection Techniques Matter for Your Career (and Your Exam)

Picture yourself scrolling through dating apps. Some profiles catch your eye immediately—clear photos, thoughtful responses, shared interests. Others make you hesitate. You're essentially doing what companies do every day when hiring: trying to predict future success based on limited information. Just as you'd probably have better luck finding a compatible partner by asking meaningful questions rather than just looking at random selfies, companies need smart, research-backed methods to identify employees who'll actually thrive in their roles.

For the EPPP, employee selection techniques represent a crucial chunk of organizational psychology. You'll need to know which methods work best, why they work, and how to combine them effectively. More importantly, understanding these techniques will help you if you ever hire staff for your own practice or evaluate workplace assessment tools. Let's break down how psychologists help organizations make better hiring decisions—and what the research actually tells us about predicting job performance.

The Core Concept: Predictors and What They Predict

Think of employee selection techniques as predictors—tools that help forecast how someone will perform in a job before they're hired. It's like weather forecasting, but instead of predicting rain, we're predicting job performance. And just like meteorologists use multiple data sources (satellite imagery, temperature readings, historical patterns), organizations use multiple techniques to make hiring decisions.

The gold standard we're predicting? Job performance. But here's the catch: not all predictors are created equal. Some work brilliantly across almost every job imaginable. Others work well only in specific situations. Let's explore each major technique, starting with the one you're probably most familiar with.

The Interview: Everyone's Favorite (Despite Its Complications)

You've probably sat through job interviews—maybe sweating in a too-formal outfit while trying to give the "perfect" answer. Organizations love interviews because they feel natural and provide face-to-face interaction. But for decades, research told us that most interviews were barely better than flipping a coin.

Here's where it gets interesting: newer research has completely changed our understanding of interview validity.

Structured vs. Unstructured Interviews

Unstructured interviews are like casual coffee conversations. The interviewer asks whatever pops into their head, jumping from topic to topic. You might get asked about your weekend plans, your college major, or some random question about moving Mount Fuji (looking at you, tech industry). Different candidates get completely different questions.

Structured interviews, by contrast, are more like standardized tests. Every candidate gets asked the same questions, derived from careful job analysis. Responses are scored using predetermined criteria. It's systematic and fair.

For years, everyone "knew" that structured interviews were far superior. But recent meta-analyses (studies that combine results from many research studies) found something surprising: both types have the same average validity coefficient of .58. That's actually pretty good! In fact, interviews are now considered the second-most valid predictor of job performance, right behind general mental ability tests.

Behavioral vs. Situational Interviews

Within structured interviews, you'll encounter two main types:

Behavioral interviews look backward. They're based on the idea that your past behavior predicts your future behavior—like checking someone's relationship history before getting serious. Questions sound like: "Tell me about a time when you had to deal with a difficult client" or "Describe a situation where you missed a deadline and how you handled it."

Situational interviews look forward. They present hypothetical scenarios and ask what you would do. For example: "Imagine a colleague consistently takes credit for your ideas in meetings. How would you handle this?"

The plot twist? A recent meta-analysis found that when both types of questions assess the same job requirements with the same applicants, situational questions were actually better predictors of job performance. This suggests that people's intentions about future behavior may predict outcomes better than their past behaviors. Think about New Year's resolutions—sometimes declaring what you'll do differently actually helps you follow through.

General Mental Ability Tests: The Heavyweight Champion

If employee selection methods were Olympic sports, general mental ability tests (also called cognitive ability tests or intelligence tests) would take home the gold medal. Research consistently shows they're the most valid predictors of job performance across virtually all jobs, performance criteria, and organizations.

Why do they work so well? Smart people tend to learn faster, solve problems more effectively, and adapt to changing circumstances. Whether you're diagnosing a complex case, managing clinic schedules, or conducting research, cognitive ability helps.

But there's a significant catch: these tests carry a higher risk of adverse impact—they may unfairly screen out qualified applicants from certain racial and ethnic minority groups. This creates a legal and ethical dilemma. The tests predict performance, but they may also perpetuate workplace inequality. Organizations must balance validity against fairness.

Personality Tests: When Conscientiousness Wins

When you're hiring, you care about more than just whether someone is smart. You want to know: Will they show up on time? Follow through on commitments? Work well with others? This is where personality tests come in.

Most organizational personality tests measure the Big Five traits:

Of these five, conscientiousness consistently emerges as the best predictor of job performance across different jobs and performance criteria. Think about it: the organized, detail-oriented, responsible colleague who actually completes their case notes on time? That's high conscientiousness in action. It matters whether you're a therapist, researcher, or administrator.

Integrity Tests: Predicting Counterproductive Behavior

Imagine you're hiring someone to work in your practice. Beyond wanting someone competent, you definitely don't want someone who'll steal from petty cash, badmouth clients, or sabotage colleagues. Integrity tests help predict whether applicants are likely to engage in counterproductive behaviors.

There are two types:

Overt integrity tests take the direct approach. They straight-up ask about attitudes toward theft and dishonesty: "Is it okay to take office supplies home?" or "Have you ever stolen from an employer?" (Surprisingly, some people actually admit to this stuff.)

Personality-based integrity tests are more subtle. They measure personality characteristics linked to counterproductive behaviors—things like impulsivity, hostility, or disregard for rules—without directly asking about misconduct.

Here's what makes integrity tests particularly valuable: they don't show adverse impact for racial or ethnic minorities, unlike cognitive ability tests. Plus, research shows they're genuinely useful. Overt tests better predict counterproductive behaviors (theft, sabotage), while personality-based tests better predict overall job performance.

The most recent meta-analysis ranked integrity tests as the fourth most valid selection method overall. Even more impressive? When you combine a general mental ability test with an integrity test, you get the greatest gain in predictive power compared to any other combination. It's like having both a smart and honest employee—the best of both worlds.

Work Samples: The "Try Before You Buy" Approach

Work samples are exactly what they sound like: you ask applicants to actually perform job-related tasks under realistic conditions. It's like test-driving a car before buying it, except you're evaluating a person's skills.

If you're hiring a therapist, you might have them conduct a mock intake interview. Hiring an administrator? Have them organize a complicated schedule or respond to difficult emails. The logic is straightforward: actual performance of job tasks should predict future performance of those same tasks.

Interestingly, work samples have dropped in validity rankings over time. Earlier research found them extremely valid—sometimes even better than cognitive ability tests. But more recent meta-analyses show lower validity. Why? Work samples used to be used primarily for manual skilled jobs (welding, carpentry, machine operation) where they worked brilliantly. Now they're increasingly used for service sector jobs where they may be less accurate.

Trainability Work Samples and Realistic Job Previews

Traditional work samples only work for experienced applicants who already have relevant skills. But what about promising candidates who lack experience? Trainability work sample tests solve this problem by incorporating training periods along with evaluation. You teach someone a task, then see how quickly they learn it. This helps identify people who'll benefit from training.

Work samples also appear in realistic job previews (RJPs)—honest presentations of both positive and negative aspects of a job. It's like a potential romantic partner being upfront about their quirks and baggage before you commit. RJPs reduce turnover by ensuring new hires have realistic expectations rather than discovering unpleasant surprises after accepting the job.

Assessment Centers: The Multi-Method Approach

Assessment centers are the comprehensive evaluation packages typically used for managerial candidates. Instead of relying on one method, they throw everything at candidates: personality tests, ability tests, structured interviews, and multiple simulations (work samples). Multiple trained raters observe and evaluate candidates across several performance dimensions.

Two classic simulations deserve mention:

In-basket exercises assess decision-making skills by presenting candidates with a pile of memos, emails, phone messages, and reports—essentially a realistic manager's inbox from hell. Candidates must prioritize items, make decisions, delegate tasks, and respond appropriately. It's like that moment when you return from vacation and face 200 unread emails, except you're being evaluated.

Leaderless group discussions evaluate leadership potential by putting a small group together to solve a job-related problem without assigning anyone as leader. Who naturally takes charge? Who facilitates good discussion? Who derails the group? It reveals leadership styles under naturalistic conditions.

Assessment centers are expensive and time-consuming, which is why they're typically reserved for high-level positions where the cost of a bad hire is substantial.

Biographical Information: Your Past Tells a Story

When biodata (biographical information blanks or BIBs) are scientifically constructed, they're surprisingly effective predictors. Unlike a typical resume, biodata forms ask multiple-choice questions about education, work history, family background, health history, interests, social relationships, and various life experiences.

The key is that items are empirically derived—chosen specifically because research shows they predict job performance. For example, research might reveal that people who participated in team sports during adolescence perform better in collaborative work environments, or that those who held leadership positions in college student organizations succeed in management roles.

Biodata predicts performance across job types, from entry-level positions to executive roles. But there's a significant downside: face validity. Some questions that scientifically predict performance don't obviously look job-related to applicants. "How many siblings do you have?" or "Did you enjoy building model airplanes as a child?" These might correlate with job success, but applicants may view them as invasive, irrelevant, or discriminatory. When people don't understand why they're being asked something, they're less likely to respond honestly or may refuse altogether.

Combining Selection Techniques: Better Together

Just as you wouldn't diagnose a client based solely on one test score or single session, organizations shouldn't hire based on one predictor alone. Multiple methods provide a more complete picture. But how do you combine information from different sources?

Compensatory Methods: Balancing Strengths and Weaknesses

Compensatory methods allow high scores on some predictors to compensate for lower scores on others. It's like being in a relationship where your partner's incredible kindness compensates for their terrible cooking.

Clinical prediction relies on human judgment. Decision makers review all the information and subjectively decide whether someone qualifies. This feels natural—we trust our instincts. But research consistently shows that statistical methods outperform human judgment. We're susceptible to biases: we favor attractive people, those who remind us of ourselves, or whoever interviewed on a good day. Our subjective impressions simply aren't as accurate as we'd like to believe.

Multiple regression is the statistical alternative. It's a formula that weights each predictor based on how much it correlates with job performance and with other predictors, then combines scores mathematically to estimate a criterion score. It removes human bias and maximizes predictive accuracy. The computer doesn't care if someone has a great smile or went to the same college you did.

Noncompensatory Methods: Minimum Standards Required

Noncompensatory methods establish minimum standards that can't be offset by strengths elsewhere. Think of them like requirements for licensure—being brilliant doesn't compensate for failing the ethics exam.

Multiple cutoff means all predictors are given to all applicants, and applicants must score above the cutoff on every single one to be considered. If you're hiring a clinical psychologist, they might need to meet minimum standards on cognitive ability, conscientiousness, emotional stability, and integrity. Excellence in one area doesn't excuse deficiency in another.

Multiple hurdles works similarly but administers predictors sequentially. Only applicants who pass each hurdle advance to the next one. This saves money when some assessments are expensive or time-consuming. Why give everyone a costly assessment center evaluation if you can first screen out clearly unqualified candidates with a quick cognitive test?

You can also combine approaches: use multiple hurdles or cutoff to identify qualified candidates, then use multiple regression among those who passed all minimums to rank-order finalists.

Common Misconceptions That Trip Up Students

Misconception #1: "Structured interviews are always better than unstructured ones." This was true in older research, but recent meta-analyses find they have equal validity (.58). Both are now recognized as highly valid predictors. Know the newer research for your exam.

Misconception #2: "Behavioral interviews are superior to situational interviews." Actually, recent research finds situational interviews (asking about hypothetical future scenarios) predict performance better than behavioral interviews (asking about past behavior) when both assess the same requirements. Intentions may predict behavior better than history.

Misconception #3: "Work samples are the most valid predictor." They used to rank extremely high, but their validity has decreased as they've been applied beyond manual skilled jobs to service positions where they're less accurate. General mental ability tests now show higher validity.

Misconception #4: "The best selection method is whichever has the highest validity." Not quite. You must also consider adverse impact, cost, applicant reactions, legal defensibility, and practical constraints. The "best" method depends on context.

Misconception #5: "Human judgment (clinical prediction) is more nuanced than statistical methods." We'd like to think so, but research consistently shows statistical methods (like multiple regression) outperform human judgment for predicting job performance. We're simply not as good at integrating complex information as we believe.

Practice Tips for Remembering This Material

Create a ranking table. Make a simple table ranking selection methods by validity from most to least valid based on the Schmidt, Oh, and Shaffer (2016) meta-analysis:

Use the acronym "I-CAN PAWS" for major selection techniques:

• Interviews
• Cognitive ability (general mental ability)
• Assessment centers
• Not sure? (placeholder)
• Personality tests
• Accuracy tests (integrity)
• Work samples
• Stories about yourself (biodata)

Okay, that's imperfect, but making your own memory devices helps encoding.

Remember combinations: When you see questions about combining methods, recall that integrity tests + general mental ability tests = greatest incremental validity. That's a high-yield exam fact.

Connect to personal experience. Think about jobs you've applied for or hiring you've participated in. Which methods did they use? How valid were they likely being? Personal connections strengthen memory.

Practice distinguishing pairs:

• Behavioral (past) vs. Situational (future) interviews
• Overt (direct) vs. Personality-based (indirect) integrity tests
• Multiple cutoff (all at once) vs. Multiple hurdles (sequential)
• Clinical (subjective) vs. Statistical (objective) prediction

Key Takeaways

• Selection techniques are predictors of future job performance, and their validity varies considerably.

• General mental ability tests are the most valid predictors across jobs but carry higher adverse impact risk for some minority groups.

• Interviews (both structured and unstructured) have a validity of .58 and are the second-most valid predictor; recent research shows situational interviews outperform behavioral interviews.

• Conscientiousness is the Big Five trait that best predicts job performance across different jobs and criteria.

• Integrity tests are highly valuable, especially when combined with general mental ability tests (greatest incremental validity of any combination).

• Work samples have decreased in validity as they've been applied beyond manual jobs to service positions.

• Assessment centers use multiple raters, multiple methods, and multiple dimensions to evaluate candidates (typically for management positions).

• Biodata can be effective but may lack face validity, causing applicant resistance.

• Compensatory methods (clinical prediction, multiple regression) allow high scores to offset low scores; statistical methods outperform human judgment.

• Noncompensatory methods (multiple cutoff, multiple hurdles) require minimum scores on all predictors; multiple hurdles is more cost-effective when assessments are expensive.

Remember, there's no single perfect selection method. The art of employee selection lies in combining multiple valid predictors while considering practical constraints, fairness, cost, and legal requirements. For the EPPP, focus on understanding the relative validity of different methods and how they can be strategically combined for better predictions. This knowledge will serve you well on exam day and in any future role where you're building a team or evaluating organizational assessment practices.