Yapılandırılmış vs Yapılandırılmamış Mülakat: 85 Yıllık Araştırma Ne Diyor

The employment interview is the most widely used selection method in the world. Virtually every organization, in every industry, in every country uses some form of interview before making a hiring decision. Yet the vast majority of these interviews are conducted in an unstructured format — and the research is unambiguous that this is one of the most consequential mistakes organizations make.

This article reviews the meta-analytic evidence spanning eight decades, from the earliest validity studies in the 1940s to the latest research on AI-assisted structured evaluation. The data tells a consistent story: structure is the single most powerful lever for improving hiring decisions.

Defining the Terms

Before examining the evidence, it is important to define what “structured” and “unstructured” mean in the context of employment interviews.

An unstructured interview is a free-flowing conversation in which the interviewer chooses questions on the fly, follows whatever tangents seem interesting, and forms an overall impression of the candidate based on the totality of the interaction. There is no predetermined framework, no standardized questions, and no scoring rubric. Each interviewer conducts a different interview, even for the same role.

A structured interview is a standardized evaluation in which:

• Questions are predetermined and based on a formal job analysis
• Every candidate for the same role is asked the same core questions, in the same order
• Responses are evaluated against behaviorally anchored rating scales (BARS)
• Each competency is scored independently before an overall assessment is formed
• Interviewers are trained on the scoring methodology

Campion, Palmer, and Campion (1997) identified 15 components that contribute to interview structure. The most impactful, in descending order of importance, are: basing questions on a job analysis, asking the same questions of every candidate, using anchored rating scales, and training interviewers.

The Meta-Analytic Evidence

The comparison between structured and unstructured interviews is one of the most thoroughly studied questions in industrial-organizational psychology. Multiple independent meta-analyses, conducted over different decades and using different methodologies, have reached the same conclusion.

Wiesner and Cronshaw (1988)

One of the earliest comprehensive meta-analyses, covering studies from the 1950s through the 1980s. Key findings:

• Structured interviews: mean corrected validity of r = 0.63
• Unstructured interviews: mean corrected validity of r = 0.20
• Structured interviews were more than three times as predictive of job performance

McDaniel, Whetzel, Schmidt, and Maurer (1994)

This meta-analysis, published in the Journal of Applied Psychology, analyzed 245 validity coefficients from 86,311 individuals:

• Structured interviews: mean corrected validity of r = 0.44
• Unstructured interviews: mean corrected validity of r = 0.33
• Situational interviews (a specific structured format): r = 0.50
• Behavioral description interviews: r = 0.48

Huffcutt and Arthur (1994)

This study specifically examined the relationship between degree of structure and validity. Rather than treating structure as a binary variable, they coded interviews on a four-level continuum from unstructured to highly structured. The finding: validity increased in a near-linear fashion with each increment of structure. Even modest increases in structure — asking a few predetermined questions alongside free-form conversation — produced measurable improvements.

Schmidt and Hunter (1998)

The landmark meta-analysis that is the most cited study in personnel selection research. Published in the Psychological Bulletin, it synthesized 85 years of research on 19 different selection methods:

Critically, Schmidt and Hunter also examined what happens when methods are combined. Structured interviews combined with general mental ability tests achieved a composite validity of r = 0.63 — the highest combined validity of any two-method pairing they studied.

Levashina, Hartwell, Morgeson, and Campion (2014)

The most comprehensive modern review, covering over 100 studies. Published in Personnel Psychology, it confirmed that the validity advantage of structured interviews holds across:

• All job types (entry-level through executive)
• All industries (healthcare, technology, finance, manufacturing, services)
• All geographies (North America, Europe, Asia-Pacific)
• Both behavioral and situational question formats

Is This Just One Study? No — It Is Converging Evidence

A common concern is that hiring teams over-index on Schmidt & Hunter (1998). That concern is healthy. In science, single-study dependence is a risk. But structured interviewing does not stand on one paper.

The key result (structure > unstructured) appears across multiple decades, journals, samples, and designs. Even when exact coefficients differ, the direction of effect is remarkably stable.

• Wiesner & Cronshaw (1988): early synthesis showing large structure advantage.
• McDaniel et al. (1994): broad meta-analysis with substantial sample size, again favoring structured formats.
• Huffcutt & Arthur (1994): near-linear gain as structure increases (not a binary cliff).
• Levashina et al. (2014): modern review confirming robustness across jobs, industries, and countries.

Practically: if one coefficient estimate is revised in future work, your hiring system does not collapse. The decision rule should be based on evidence convergence, not one citation.

Why Structure Works: The Psychological Mechanisms

The validity advantage of structured interviews is not accidental. It results from specific psychological mechanisms:

1. Noise Reduction

Kahneman, Sibony, and Sunstein (2021) identified “noise” — unwanted variability between evaluators — as a primary source of decision error. Structured interviews reduce all three types of noise:

• Level noise is reduced by anchored rating scales that calibrate what a “3” versus a “5” means
• Pattern noise is reduced by standardized questions that ensure all evaluators assess the same competencies
• Occasion noise is reduced by the discipline of the format, which constrains the influence of mood, fatigue, and context effects

2. Bias Mitigation

Structured interviews limit the influence of cognitive biases that contaminate unstructured evaluations:

• First-impression bias — When interviewers must evaluate specific competencies against specific evidence, the influence of snap judgments diminishes (Barrick, Swider & Stewart, 2010)
• Similarity bias — Standardized criteria reduce the weight given to irrelevant demographic similarities (Rivera, 2012)
• Halo/horn effects — Independent scoring of each competency prevents a single strong or weak response from coloring the entire evaluation

3. Information Quality

Structured questions, particularly behavioral questions (“Tell me about a time when…”), elicit more job-relevant information than open-ended conversation. Janz (1982) demonstrated that behavioral description interviews produce richer, more specific, and more verifiable candidate responses than unstructured formats.

4. Decision Discipline

The requirement to score each competency independently before forming an overall impression implements what Kahneman (2021) calls “delayed holistic judgment.” This forces evaluators to consider all dimensions of a candidate’s performance before reaching a conclusion — rather than allowing a single dominant impression to drive the decision.

Why Organizations Resist Structure

Given the overwhelming evidence, why do most organizations still default to unstructured interviews? The research identifies several explanations:

The Illusion of Insight

Interviewers who conduct unstructured interviews are more confident in their predictions than those who rely on structured data — despite making worse predictions.

— Dana, Dawes & Peterson (2013), Judgment and Decision Making

The conversational format creates a subjective sense of understanding the candidate that is not supported by outcomes. This illusion is self-reinforcing: interviewers believe they are skilled judges of character and resist formats that constrain their autonomy.

Preference for Flexibility

Hiring managers often prefer the flexibility to “go where the conversation takes them” and believe that rigidity will cause them to miss important signals. The meta-analytic evidence directly contradicts this belief: the signals that unstructured interviewers follow are, on average, less predictive than those captured by structured formats.

Training and Habit

Most hiring managers have never been trained to conduct structured interviews. Their mental model of an interview is a conversation — because every interview they have ever conducted or experienced has been one. Changing this default requires both training and tools.

Implementation Cost

Historically, structured hiring required significant upfront investment: job analysis, competency mapping, question development, rubric design, and interviewer training. For organizations without I/O psychology expertise, this was a prohibitive barrier. AI has largely eliminated this barrier, generating competency frameworks, questions, and rubrics in minutes rather than weeks.

The Semi-Structured Middle Ground

It is worth noting that structure is not binary. Huffcutt and Arthur’s (1994) finding — that validity increases near-linearly with structure — implies that any movement toward structure improves outcomes. Organizations that are not ready for fully structured interviews can start with a semi-structured approach:

• Define three to four core questions that every candidate must answer
• Allow limited follow-up flexibility within each question area
• Use a simple rating scale (even 1–3) for each core area
• Score independently before sharing impressions

Even these modest structural elements will measurably improve inter-rater agreement and predictive validity.

What Critics Are Right About (And How to Handle It)

Critics of legacy meta-analyses raise valid points: labor markets change, jobs evolve, and measurement assumptions can shift across decades. That is exactly why evidence-based teams should combine external research with local validation.

• Use global evidence for direction: structured methods are the safer default.
• Use internal data for calibration: track validity, pass-through rates, and quality-of-hire in your own context.
• Audit quarterly: verify consistency, adverse impact, and interviewer drift.

This approach makes your process resilient: even if academic effect sizes are updated, your decision system remains grounded in observed outcomes.

The Impact on Fairness and Adverse Impact

Structure does not only improve prediction — it also improves fairness. Huffcutt and Arthur (1994) found that structured interviews show less adverse impact against racial and ethnic minority groups than unstructured interviews. The mechanism is straightforward: when evaluation criteria are standardized and evidence-based, the influence of irrelevant demographic characteristics diminishes.

Bohnet (2016) extended this finding to gender, demonstrating that structured evaluation environments reduce the gender gap in hiring outcomes — not by lowering standards, but by ensuring that the same standards are applied to everyone.

This fairness advantage has regulatory implications. Under the EU AI Act, organizations using AI in hiring must demonstrate that their systems do not produce discriminatory outcomes. Structured interviews provide the documentary foundation for this demonstration.

The Candidate Experience Advantage

A common misconception is that candidates prefer unstructured, conversational interviews. The research suggests otherwise. Chapman and Zweig (2005) found that candidates rate structured interviews as more fair, more job-related, and more professional than unstructured ones. Candidates perceive that they are being evaluated on relevant criteria rather than arbitrary personal chemistry — and they are right.

In competitive talent markets, this perception matters. The candidate experience during the interview process directly influences offer acceptance rates and employer brand perception. Structure does not sacrifice candidate experience — it enhances it.

Summary of the Evidence

The Bottom Line

The strongest scientific position is not “believe one famous paper.” It is: trust converging evidence and continuously validate locally. On that standard, structured interviewing remains one of the most defensible upgrades a hiring team can make.

AI now makes implementation practical at scale, but governance still matters. Use structure, require evidence for every score, monitor outcomes over time, and keep your method reviewable. That combination is far more robust than intuition-led interviewing and far less fragile than single-study thinking.

References

• Barrick, M. R., Swider, B. W., & Stewart, G. L. (2010). Initial evaluations in the interview. Journal of Applied Psychology, 95(6), 1163–1172.
• Bohnet, I. (2016). What Works: Gender Equality by Design. Harvard University Press.
• Campion, M. A., Palmer, D. K., & Campion, J. E. (1997). A review of structure in the selection interview. Personnel Psychology, 50(3), 655–702.
• Chapman, D. S., & Zweig, D. I. (2005). Developing a nomological network for interview structure. Personnel Psychology, 58(3), 673–702.
• Dana, J., Dawes, R., & Peterson, N. (2013). Belief in the unstructured interview. Judgment and Decision Making, 8(5), 512–520.
• Huffcutt, A. I., & Arthur, W. (1994). Hunter and Hunter (1984) revisited. Journal of Applied Psychology, 79(2), 184–190.
• Janz, T. (1982). Initial comparisons of patterned behavior description interviews versus unstructured interviews. Journal of Applied Psychology, 67(5), 577–580.
• Kahneman, D., Sibony, O., & Sunstein, C. R. (2021). Noise: A Flaw in Human Judgment. Little, Brown Spark.
• Levashina, J., Hartwell, C. J., Morgeson, F. P., & Campion, M. A. (2014). The structured employment interview. Personnel Psychology, 67(1), 241–293.
• McDaniel, M. A., Whetzel, D. L., Schmidt, F. L., & Maurer, S. D. (1994). The validity of employment interviews: A comprehensive review and meta-analysis. Journal of Applied Psychology, 79(4), 599–616.
• Rivera, L. A. (2012). Hiring as cultural matching. American Sociological Review, 77(6), 999–1022.
• Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods. Psychological Bulletin, 124(2), 262–274.
• Wiesner, W. H., & Cronshaw, S. F. (1988). A meta-analytic investigation of the impact of interview format and degree of structure on the validity of the employment interview. Journal of Occupational Psychology, 61(4), 275–290.