The Return of Standardized Testing in Tech Hiring

The Resurgence of SAT Scores in Technical Hiring

Some tech startups are returning to requesting SAT scores on job applications as a means of verifying raw, non-AI-assisted cognitive ability. This trend emerges as generative AI makes traditional screening tools—such as resumes, cover letters, and take-home assignments—easier to automate, potentially masking a candidate's actual "raw horsepower."

While standardized tests are often viewed as outdated, proponents argue they provide an objective benchmark of foundational fluency that is harder to fake than a polished, AI-generated portfolio. However, using decades-old scores introduces significant noise, as these tests reflect a candidate's status as a teenager rather than their current professional capability.

Evidence-Based Predictors of Job Performance

Not all hiring assessments are created equal. Research indicates that a combination of specific methods yields the highest predictive validity for professional success in knowledge work.

High-Validity Methods

• Structured Interviews: These are currently considered among the strongest predictors of performance. To be effective, they require consistent questions, a standardized scoring rubric, and trained interviewers to minimize bias.
• Work Sample and Job Knowledge Tests: Having candidates perform a representative slice of the actual job (e.g., a trial period or a specific technical task) provides a direct observation of competency.
• Cognitive Ability Tests: These remain strong predictors, particularly for roles with steep learning curves or heavy training requirements, though they are generally less predictive than structured interviews.
• Assessment Centers: Utilizing simulations and multiple assessors, these are highly effective for predicting managerial aptitude but are often too expensive and time-consuming for general hiring.

Low-Validity Methods

• Unstructured Interviews: Casual "chats" are among the least effective predictors due to their susceptibility to the halo effect, impression management, and "similar-to-me" bias.
• Brain Teasers: Once popularized by companies like Google and Microsoft, these are now widely regarded as a waste of time that fails to predict job performance.
• Personality Tests: Tools like Myers-Briggs are generally considered ineffective for predicting professional success.

Challenges of Using SAT Scores as a Metric

Using SAT scores as a hiring filter presents several technical and ethical challenges that can undermine their utility as a data point:

• Score Inconsistency: SAT scoring scales have changed multiple times (shifting between 1600 and 2400 maximums), making it difficult to compare candidates from different cohorts.
• Environmental Noise: Scores are heavily influenced by the conditions under which the test was taken, including access to tutoring, home stability, and the candidate's mental state at age 17.
• Limited Scope: Standardized tests may proxy for general intelligence or curiosity but fail to measure essential professional traits such as reliability, collaboration, and communication skills.

Adapting Hiring for the AI Era

To combat the "perfect, phony all-star" effect created by AI-generated applications, hiring managers are encouraged to move toward verifiable, proctored, and current assessments.

Recommended AI-Resistant Alternatives

• Live, Timed Exercises: Conducting technical tests in person or via proctored remote sessions prevents candidates from using LLMs to generate answers in real-time.
• Defensive Work Reviews: If a take-home project is used, the subsequent interview should focus on an "interrogation" of the work, requiring the candidate to defend their logic and explain their process to ensure the work is their own.
• In-Person Whiteboarding: Returning to physical whiteboarding allows interviewers to observe the candidate's thought process in real-time.
• Live Roleplay: For GTM (Go-To-Market) or sales roles, live simulations are the most effective way to gauge interpersonal and communication skills.

Community Perspectives on Standardized Testing

Discussion among industry professionals reveals a divide on the utility of these tests:

"SAT scores correlate to IQ at .81, and IQ is one of the few things that strongly correlates to knowledge work performance positively. There is probably a lot of alpha from knowing candidates SAT scores."

Conversely, others argue that such requirements can act as a cultural filter or a deterrent for highly qualified candidates:

"I had a friend with an MS show up for first-day of work for a job that asked for SAT scores... He realized that he didn't want to work for a company that was this serious about SAT scores when hiring someone with a post-graduate degree."