๐Ÿ‡ฎ๐Ÿ‡ฉ Indonesia: SNBT UTBK-IRT Weighting Predictor

Estimate your SNBT-UTBK score on the 200โ€“1000 scale used by SNPMB. Item Response Theory weights harder items more heavily โ€” raw hits and a difficulty profile are required across all 7 subtests.

Skip to Results

IRT assigns variable points per item. "Mostly Hard" applies the highest difficulty multiplier (โ‰ˆ 1.15ร—), "Mostly Easy" the lowest (โ‰ˆ 0.85ร—).

Penalaran Umum (/30)
Pengetahuan & Pemahaman Umum (/20)
Pemahaman Bacaan & Menulis (/20)
Pengetahuan Kuantitatif (/20)
Literasi Bahasa Indonesia (/30)
Literasi Bahasa Inggris (/20)
Penalaran Matematika (/20)

SNBT-UTBK 2026 totals 160 items across 7 subtests in 195 minutes. Blank and wrong answers both score 0 โ€” no penalty.

Typical cutoffs: ITB/UI Kedokteran โ‰ˆ 720โ€“780, Teknik Informatika top PTN โ‰ˆ 680โ€“730, Saintek mid-tier โ‰ˆ 600โ€“650.

Predicted UTBK Score (200โ€“1000)

--
Enter raw hits for all 7 subtests
TPS Composite --
Literasi Composite --
Penalaran Matematika --
IRT Difficulty Multiplier 1.00ร—
Total Raw Hits -- / 160
Score vs Cutoff --
SNPMB does not publish exact IRT discrimination parameters โ€” this tool models the public scale (200โ€“1000) with a difficulty multiplier proxy.

How It Works

1. Item Response Theory (IRT)

SNPMB scores UTBK with IRT rather than a simple percentage. Each item carries a difficulty parameter b calibrated from how the national cohort answered it. Correct answers on hard items contribute more, easy items contribute less, and wrong/blank answers all score zero โ€” there is no negative marking.

Because the exact b values are confidential, this tool uses your self-reported difficulty profile as a proxy multiplier (Easy โ‰ˆ 0.85, Mixed โ‰ˆ 1.00, Hard โ‰ˆ 1.15).

2. The 7 Subtests (160 items)

TPS: Penalaran Umum (30), PPU (20), PBM (20), PK (20). Literasi: Bahasa Indonesia (30), Bahasa Inggris (20), Penalaran Matematika (20). Total 160 items in 195 minutes.

Each subtest is scored on its own 200โ€“1000 sub-scale, then the seven sub-scale scores are averaged equally to produce the overall UTBK score reported to PTNs.

3. Subscale Mapping (200โ€“1000)

The accepted public approximation for each subtest is: score = 200 + (hits รท items) ร— 800 ร— IRT_mult, clamped to [200, 1000]. The composite is the unweighted mean of all seven subscales.

This avoids a common trap: simple percentage calculators report e.g. "80%" โ€” but UTBK never reports percentages. PTNs only see the 200โ€“1000 score.

4. Adversarial Boundary Trap

Zero hits in a single subtest still floor that subscale at 200 โ€” but they drag your composite. A candidate with 160/160 raw hits on the Easy profile (multiplier 0.85) maxes out at ~880, while 130/160 on the Hard profile (1.15) lands roughly 840โ€“936 depending on spread โ€” often matching or exceeding the perfect Easy run. Difficulty selection matters as much as hit count.

Inputs out of range are clamped to [0, item count]; missing fields are treated as zero hits, which transparently surfaces under-performance instead of masking it.

Primary Regulatory Sources

  • Official SNPMB (Seleksi Nasional Penerimaan Mahasiswa Baru) portal: snpmb.bppp.kemdikbud.go.id โ€” administered by Balai Pengelolaan Pengujian Pendidikan, Kemdikbudristek.
  • SNPMB Framework for UTBK: snpmb.id โ€” official rules, materi, and timeline for SNBT, SNBP, and UTBK 2026.

Latest Articles

Fresh guides and advice for studying abroad.

View all →
System Share
Copy Link
More Share Options