Failed Saber 11? Here's Your Recovery Plan

Pro tip: A retake does not overwrite anything. Each sitting produces its own report, and receiving institutions decide whether to consider your best or your most recent result. For the annual classification of educational establishments, ICFES counts only the first time a candidate sat the exam as a grade 11 or grade 26 student, so later attempts do not affect a school's figures. ICFES publishes no ranked list of individual best results at all.

• Treating the five pruebas as equally weighted
  English carries 1 of 13 weighting points in the Global Score while Lectura Critica, Matematicas, Ciencias Naturales and Sociales y Ciudadanas each carry 3. An hour spent on Ciencias Naturales moves the Global Score three times as far as an hour spent on English.
• Studying Ciencias Naturales as three separate school subjects
  ICFES publishes a competency by component matrix. Indagacion alone is 40 percent of the prueba and cuts across biology, physics and chemistry. Practising experimental design and data interpretation pays more than revising a chapter of any single subject.
• Preparing Lectura Critica only with literary texts
  Continuous literary texts are about 24 percent of the prueba. Informative philosophical texts are about 30 percent and informative non-philosophical texts another 30 percent, with discontinuous texts making up about 16 percent.
• Drilling computation for Matematicas
  The largest competency is formulacion y ejecucion at 43 percent, followed by interpretacion y representacion at 34 percent and argumentacion at 23 percent. Items ask you to choose or justify a procedure far more often than to grind through one.
• Ignoring the socioeconomic and school climate questionnaires
  They are 48 of the 302 items in the standard booklet and consume real minutes in both sessions, but produce no academic score. Answer them briskly so the time goes to the pruebas that count.
• Choosing the standard booklet when a reasonable-adjustment booklet is available
  ICFES states plainly that a candidate with a disability who chooses the standard booklet answers more questions and therefore has less time per question. The supports requested at registration apply either way, so the booklet choice is a separate decision worth making deliberately.
• Registering in the extraordinary window by default
  For 2026 the extraordinary tarifa is roughly 50 percent higher across every category: COP 108,000 instead of COP 72,000 for a public school student, and COP 215,000 instead of COP 143,000 for a graduated bachiller.
• Assuming a raw percentage of correct answers converts to a score
  Each prueba is scored with a three-parameter logistic model in which the probability of a correct response depends on the item's characteristics as well as the candidate's ability. Two candidates with the same number of correct answers can receive different scores.
• Filling answer bubbles carelessly
  ICFES instructs candidates to completely fill the circle corresponding to the chosen option. A partial mark on a paper answer sheet is a lost item on a test where every item is worth the same before scaling.
• Reading only the Global Score on the results report
  The report gives a 0 to 100 score, a percentile, a performance level and a written description for each prueba, plus the Global Score percentile. The per-prueba percentiles tell you where to work; the Global Score alone does not.

• Computer-based national exam
• 5 areas: Critical Reading, Math, Science, Social, English
• 500 max score per area
• Required for all university applications
• ICFES Interactivo for practice
• Results used for scholarships