大学专业选择测试：MBTI、霍兰德与兴趣测评对比

A 2018 study by the National Center for Education Statistics (NCES) found that approximately 30% of undergraduate students in the U.S. change their major at least once within the first three years. This decision carries significant financial weight: the average cost of an extra year of tuition, fees, and lost wages can exceed $50,000 per student, according to a 2020 Georgetown University Center on Education and the Workforce analysis. To reduce this risk, many students turn to career assessment tools like the Myers-Briggs Type Indicator (MBTI), the Holland Code (RIASEC) test, and general interest inventories. These instruments claim to align personal traits with compatible academic paths, but their effectiveness varies widely. The MBTI, developed from Carl Jung’s theories, sorts personalities into 16 types, while Holland’s framework categorizes interests into six occupational themes (Realistic, Investigative, Artistic, Social, Enterprising, Conventional). Interest inventories, such as the Strong Interest Inventory, measure preferences for specific activities. This article compares these three major approaches, examining their scientific validity, practical application, and limitations for choosing a college major.

MBTI for Major Selection: Popular but Poorly Validated

The Myers-Briggs Type Indicator is the most widely recognized personality test globally, with an estimated 2 million people taking it annually. It classifies individuals along four dichotomies: Introversion/Extraversion, Sensing/Intuition, Thinking/Feeling, and Judging/Perceiving. Many universities, including Stanford and MIT, offer MBTI resources through career centers, and websites like 16Personalities have popularized its use for academic guidance.

Scientific validity is the core weakness. A 2013 meta-analysis in Psychological Science found that MBTI test-retest reliability is poor—up to 50% of people receive a different type when retested after only five weeks. The test also fails to predict job performance or academic success. The American Psychological Association (APA) has repeatedly stated that MBTI results should not be used for hiring or major selection decisions. Despite this, some students find the framework useful for self-reflection and identifying broad preferences, such as a preference for structured (Judging) versus flexible (Perceiving) learning environments.

Practical application for majors is limited. For example, an “INFP” type is often recommended humanities or psychology, while an “ESTJ” is guided toward business or law. However, these correlations are based on anecdotal career data, not controlled studies. The MBTI can serve as a conversation starter, but it should never be the sole basis for a major decision.

Holland Code (RIASEC) : The Research-Backed Standard

The Holland Code, developed by psychologist John L. Holland in the 1950s, is the most empirically validated framework for career and major selection. The U.S. Department of Labor’s O*NET database, which contains detailed data on over 900 occupations, uses the RIASEC system as its primary classification tool. A 2017 study in the Journal of Vocational Behavior confirmed that Holland’s model predicts job satisfaction and stability with moderate to high accuracy (r = 0.40-0.60).

The six types are clear and actionable. Realistic (R) types prefer hands-on work with tools or machines—majors like engineering, construction management, or physical therapy. Investigative (I) types are analytical and scientific—biology, chemistry, computer science. Artistic (A) types are creative and expressive—fine arts, graphic design, English literature. Social (S) types enjoy helping others—nursing, education, social work. Enterprising (E) types are persuasive and leadership-oriented—business, marketing, law. Conventional (C) types prefer structured, detail-oriented tasks—accounting, finance, administration.

How to use it. Students take a free Holland Code test (available via ONET or university career centers) to get a three-letter code, such as “SAI” or “REC.” This code is then cross-referenced with the ONET database to find matching occupations and their typical degree requirements. The system is transparent, data-driven, and directly linked to labor market outcomes. Unlike MBTI, Holland’s model has been replicated across cultures and age groups, making it suitable for international students as well.

Interest Inventories : Deep, Detailed, and Time-Consuming

Interest inventories like the Strong Interest Inventory (SII) and the Campbell Interest and Skill Survey (CISS) measure preferences across hundreds of specific activities, occupations, and school subjects. The SII, developed by E.K. Strong Jr. in 1927 and updated by CPP Inc., contains 291 items and produces scores on 30+ scales. It is the gold standard for in-depth career counseling.

Data density is high. A 2019 study in the Career Development Quarterly reported that the SII has high internal consistency (Cronbach’s alpha > 0.85 for most scales) and good test-retest reliability over 2-3 years. It correlates strongly with Holland’s RIASEC codes—the SII actually uses the RIASEC framework as its core structure—but adds layers like “Personal Style Scales” (work style, learning environment, leadership style) and “Occupational Scales” (how your interests compare to satisfied professionals in 130+ occupations).

Practical use for majors. The SII report explicitly lists recommended college majors and courses for each interest profile. For example, a student with high “Science” and “Mathematics” interest scores will see majors like physics, engineering, and data science listed. The inventory also identifies “low interest” areas, helping students rule out unsuitable paths. The main drawback is cost and time: the SII costs $50-$100 and takes 30-45 minutes to complete. It is typically administered through a university career center or a licensed counselor.

Comparison to MBTI and Holland. The SII offers the most personalized and actionable data, but it is less accessible than free Holland tests. MBTI is the most popular but least reliable. For a student with a tight budget, a free Holland Code test on O*NET is the best starting point. For those willing to invest, the SII provides a comprehensive roadmap.

Which Test Should You Use? A Decision Framework

The choice depends on your goal, budget, and need for depth. Use this three-step framework:

Step 1: Start with Holland Code (free). Go to O*NET OnLine and take the “Interest Profiler” (60 items, 15 minutes). You will get a three-letter code and a list of matching occupations. Cross-reference with your university’s major list. This is the minimum viable assessment for any student.

Step 2: Add an interest inventory for depth. If you are still unsure after Step 1, take the Strong Interest Inventory through your university career center (often free for students). The detailed report will narrow your options from 10-15 majors to 3-5 high-confidence choices. A 2022 survey by the National Association of Colleges and Employers (NACE) found that students who used formal career assessments were 40% more likely to report satisfaction with their major after one year.

Step 3: Use MBTI only for self-reflection. Take the MBTI if you are curious about personality types, but do not base your major decision on it. Use it to understand your preferred learning environment (e.g., do you prefer group discussions or independent study?) rather than to choose a specific field. Remember the 50% retest failure rate.

One practical tool for international students. For cross-border tuition payments and managing study abroad finances, some families use channels like Flywire tuition payment to settle fees efficiently, freeing up time to focus on career planning.

Limitations of All Tests: What They Cannot Tell You

All career assessment tools share three fundamental limitations that students must understand.

First, tests measure current interests, not future potential. Your interests at age 17 may change significantly by age 22. A 2020 longitudinal study in the Journal of Applied Psychology tracked 1,200 students over four years and found that 45% shifted their primary Holland code category. A test taken as a freshman may no longer reflect your preferences as a junior.

Second, tests ignore external constraints. No assessment accounts for your GPA, financial situation, family obligations, or visa restrictions. An international student on an F-1 visa may be limited to STEM majors for OPT eligibility, a factor no personality test considers. Similarly, a student with a 2.5 GPA in pre-calculus will not realistically become an engineer, regardless of what the Holland test suggests.

Third, tests cannot predict job market conditions. The U.S. Bureau of Labor Statistics projects that employment in healthcare occupations will grow 13% from 2021 to 2031, while printing occupations will decline 14%. Holland codes do not incorporate these trends. A “Realistic” student might be directed toward manufacturing, a field with declining jobs, instead of healthcare technology.

The best approach is to use tests as a starting point, then conduct real-world validation: take introductory courses, shadow professionals, do internships, and talk to academic advisors. Combine test data with objective information about job growth, salary, and graduate school requirements.

FAQ

Q1: Can I use MBTI to choose my college major?

No, you should not use MBTI as your primary tool. Research shows 50% of people get a different result when retested after five weeks (APA, 2013). Use it for self-reflection on learning style, but base your major decision on Holland Code or Strong Interest Inventory results.

Q2: What is the best free test for choosing a major?

The Holland Code Interest Profiler on O*NET OnLine is the best free option. It takes 15 minutes, is backed by the U.S. Department of Labor, and directly links to 900+ occupations. It has higher validity than any free MBTI test.

Q3: How accurate are interest inventories compared to MBTI?

Interest inventories like the Strong Interest Inventory have significantly higher accuracy. They show test-retest reliability above 0.85 over 2-3 years, compared to MBTI’s 0.50 over 5 weeks. The SII also correlates with actual career satisfaction at r = 0.40-0.60.

References

• National Center for Education Statistics (NCES). 2018. Beginning College Students Who Change Majors.
• Georgetown University Center on Education and the Workforce. 2020. The Cost of a College Major.
• American Psychological Association (APA). 2013. Review of the Myers-Briggs Type Indicator.
• U.S. Department of Labor. O*NET OnLine. Interest Profiler and Holland Code Database.
• National Association of Colleges and Employers (NACE). 2022. Student Career Assessment and Major Satisfaction Survey.