Error logs and spaced review

Lesson 10 showed how to review practice exams. This lesson turns that review into a system. An error log captures every mistake and near miss so that patterns become visible. Spaced review makes sure that what you learn from those mistakes is still in your memory on exam day. Together they make the hours you spend on practice questions count for far more.

The underlying idea comes from research on learning. Two findings are especially well supported: actively retrieving information, as in answering questions or recalling a concept from memory, strengthens memory more than rereading it; and spreading study over time, with growing gaps between reviews, produces more lasting learning than cramming. Both are easy to apply to PMP preparation.

What an error log is

An error log is a simple record of every practice question you got wrong or got right by guessing. It can be a spreadsheet, a notebook or a set of digital flashcards. For each entry, record enough to learn from it later without needing the original question.

The most important column is the last one. A lesson written as a rule you can apply, such as when a risk with a planned response occurs, implement the response first, will help with many future questions. A note that simply records the right letter will not.

Finding patterns

After a few practice exams the log will contain dozens of entries. Group them by error type and topic, and count. A Pareto view, sorted from most to least frequent, shows where a little effort will remove a lot of errors.

# Summarizing Priya's error log after three practice exams, then scheduling reviews.
from collections import Counter
log = (["knowledge: procurement"] * 6 + ["knowledge: compliance"] * 9 + ["misread stem"] * 11 +
       ["wrong mindset: escalated too soon"] * 8 + ["calculation slip"] * 4 +
       ["knowledge: agile roles"] * 5 + ["time pressure"] * 3)
counts = Counter(log)
total = sum(counts.values())
running = 0
print(f"{'Cause':36}{'Count':>6}{'Cum %':>7}")
for cause, n in counts.most_common():
    running += n
    print(f"{cause:36}{n:>6}{running / total:>7.0%}")
print()
# Expanding review intervals (days) for one error-log card, in the spirit of a Leitner system.
intervals, day = [1, 3, 7, 14], 0
for i in intervals:
    day += i
    print(f"review on day {day}")

Cause                                Count  Cum %
misread stem                            11    24%
knowledge: compliance                    9    43%
wrong mindset: escalated too soon        8    61%
knowledge: procurement                   6    74%
knowledge: agile roles                   5    85%
calculation slip                         4    93%
time pressure                            3   100%

review on day 1
review on day 4
review on day 11
review on day 25

Priya's log after three practice exams has 46 entries. The largest group is misreading the stem, nearly a quarter of all errors. Next come knowledge gaps in compliance and a mindset error, escalating too soon. Those three causes account for about 61 % of her mistakes. Her plan follows the evidence: she adopts a deliberate reading routine for every stem, works through the compliance lessons of PMP-150 again with focused questions, and reviews the decision ladder from lesson 3 so that she assesses and acts within authority before escalating.

The biggest single cause is not a knowledge gap at all. Many candidates assume that more reading will fix their scores, when their real problem is how they read questions or how they reason about them. An error log makes that visible.

Spaced review

Lessons from the error log fade unless they are reviewed. Spaced review schedules each item for review at growing intervals: for example the next day, then three days later, then a week later, then two weeks later. If you recall the lesson easily, the interval grows; if you struggle, it shrinks back. This is the idea behind the Leitner box system and most flashcard apps.

A practical approach for PMP preparation is to turn each error-log lesson into a question-and-answer card. The front states the situation in a sentence; the back gives the principle. Review cards for 10 to 15 minutes a day, rather than in long, occasional sessions. Cards for formulas, such as the earned value set, work the same way.

Retrieval practice beyond cards

Retrieval practice means pulling knowledge out of memory rather than looking at it again. Practice questions are retrieval practice, which is one reason they work so well. Other ways to retrieve include explaining a concept aloud as if teaching it, writing the earned value formulas from memory on a blank page, or drawing the decision ladder without looking. These take a few minutes and reveal gaps immediately.

Interleaving is a related technique: mixing topics within a study session rather than studying one topic at a time. It feels harder, but it builds the skill the exam requires, which is recognizing what kind of question you face. Mixed practice sets, recommended throughout this course, are interleaved by design.

Managing the volume

A busy candidate can log many errors in a single full-length exam. If the log becomes unmanageable, it stops being used. Keep it lean:

• Log every wrong answer and every right-but-unsure answer from full-length exams; from short quizzes, log only new lessons.
• Merge entries that teach the same lesson, keeping a count of how often it appears.
• Retire cards you have recalled easily on three consecutive reviews, but bring them back if the same error reappears.
• Spend most review time on the top causes in your Pareto view.

If you study with colleagues, a shared error log can help, but keep personal reasoning in your own log. Several PMI chapters in Canada run study groups where members compare reasoning on practice questions; explaining your thinking to others is excellent retrieval practice.

Building an error log step by step

Setting up an error log takes about fifteen minutes. If you use a spreadsheet, create the columns from the table above, add a column for the review date, and freeze the header row. If you use a flashcard app, create one deck for situational lessons and one for formulas and terms, and use tags for domain and error type so that you can filter later. Whatever the tool, keep the entry format consistent, because consistency makes patterns easy to see.

Fill in the log during your review sessions, not during practice tests. While taking a test, simply flag questions and note whether you were sure; record the details afterwards when you read the explanations. After each full-length exam, update the counts by error type and topic and look at the top three. Those three causes should shape the next week of study.

Over time, the log tells a story. Early in preparation, most entries are knowledge gaps. As you study, knowledge errors fall and reading and mindset errors become a larger share. Late in preparation, the remaining errors are often close calls between two good options. That progression is normal, and it means the kind of practice you need changes: from learning content, to reading carefully, to sharpening judgement on fine distinctions.

A week of spaced review in practice

A sample week shows how little time spaced review needs. Each morning, Priya spends ten minutes on her review cards over coffee: new cards from the last review session, plus the cards scheduled for that day. On two evenings she adds a twenty-minute retrieval session, writing the formulas from memory and explaining two error-log lessons aloud. At the weekend she takes a full-length exam or a mixed set and logs new lessons. The total extra time is under two hours a week, yet by exam day each important lesson has been retrieved four or five times at growing intervals.