LET Elementary Assessment of Learning — Statistics, Grading and Interpreting Assessment ResultsRevision Notes

Formulas

Exam Tips

• LET computation: Always sort scores before computing the median. Unsorted data is a deliberate trap.
• If an item asks which measure to report for a skewed distribution, the answer is always the MEDIAN.
• Memorize: In a normal distribution, Mean = Median = Mode. Any deviation from this signals skewness.
• If an item gives you a very high or very low outlier score and asks which measure is most affected, the answer is the MEAN.

Key Points

• A measure of central tendency summarizes a whole distribution with a single representative value.
• The MEAN is the arithmetic average: Sum of all scores divided by the number of scores (n). It uses every score, making it the most powerful but also the most sensitive to outliers.
• The MEDIAN is the middle score when data are arranged in order. For an even number of scores, it is the average of the two middle scores. The median equals the 50th percentile (P50).
• The MODE is the score that appears most frequently. A distribution can be unimodal (one mode), bimodal (two modes), or have no mode at all.
• Use the MEAN for roughly symmetric distributions with no extreme scores.
• Use the MEDIAN when the distribution is skewed or contains outliers — it is resistant to extreme values.
• Use the MODE for nominal (categorical) data or when you need the most typical/popular value.
• In a normal (symmetric) distribution: Mean = Median = Mode.
• Worked Example: Scores of 7 pupils — 82, 85, 85, 88, 90, 92, 94. Sum = 616. Mean = 616 ÷ 7 = 88. Median = 88 (4th score in order). Mode = 85 (appears twice).

Definitions

Section Title

Measures of Central Tendency

Common Mistakes

• Confusing the mode with the median — the mode is the MOST FREQUENT score, not the middle score.
• Forgetting to ARRANGE scores in order before identifying the median.
• Using the mean to summarize a skewed distribution — always ask first: are there extreme scores?
• Thinking there can only be one mode — a distribution can be bimodal or even multimodal.
• Assuming the mean, median, and mode are always different — in a perfectly symmetrical (normal) distribution, they are all equal.

Formulas

Exam Tips

• LET trap: If asked what a large SD implies about the class, answer HETEROGENEOUS (varied scores); small SD = HOMOGENEOUS.
• Know the conceptual steps for SD even if full computation is not required: deviations → square → average → square root.
• IQR is always about the middle 50% of the distribution — link it to Q1 and Q3.
• The SD is the building block for z-scores and the normal curve — master it here first.

Key Points

• Variability (dispersion) measures how spread out scores are around the center.
• Two classes can have the same mean but very different score distributions — variability captures that difference.
• The RANGE is the simplest measure: highest score minus lowest score. It is quick but unstable because it depends on only two extreme values.
• The STANDARD DEVIATION (SD) is the most reliable and widely used measure of variability. Conceptually, it is the average distance of all scores from the mean.
• A SMALL SD means scores cluster closely around the mean — the class is HOMOGENEOUS.
• A LARGE SD means scores are widely spread — the class is HETEROGENEOUS.
• The VARIANCE is the SD squared. The SD is the square root of the variance.
• The INTERQUARTILE RANGE (IQR = Q3 - Q1) captures the spread of the middle 50% of scores and, like the median, is resistant to outliers.
• Worked Example: Scores 4, 6, 8, 10, 12. Mean = 8. Deviations: -4, -2, 0, +2, +4. Squared deviations: 16, 4, 0, 4, 16. Sum = 40. Variance = 40/5 = 8. SD = √8 ≈ 2.83.

Definitions

Section Title

Measures of Variability

Common Mistakes

• Forgetting to SQUARE the deviations before summing them — unsquared deviations always sum to zero.
• Confusing SD with variance — variance is SD squared; SD is the square root of variance.
• Thinking a large range automatically means a large SD — two outliers can inflate the range while most scores remain tightly clustered.
• Mixing up homogeneous (small SD) with heterogeneous (large SD).
• Using range as the sole measure of variability — range is crude and unstable.

Formulas

Exam Tips

• Whenever a stem says 'percentile rank of X,' the correct interpretation is: performed better than X% of the group.
• Memorize: Q1=P25, Q2=P50=Median=D5, Q3=P75. One item almost always tests this chain.
• IQR = Q3 − Q1 always covers the MIDDLE 50% — link this to box-and-whisker plots conceptually.

Key Points

• Percentiles divide an ordered distribution into 100 equal parts. A percentile RANK of 90 means the student scored as well as or better than 90% of the group — it does NOT mean the student got 90% of items correct.
• Quartiles divide the distribution into FOUR equal parts: Q1 (P25), Q2 (P50), Q3 (P75).
• Q2 = P50 = D5 = the MEDIAN. This chain of equalities is a classic LET item.
• Deciles divide the distribution into TEN equal parts (D1 through D9). D5 equals the median.
• The Interquartile Range (IQR = Q3 − Q1) describes the spread of the middle 50% of scores.
• Percentiles are RELATIVE (norm-referenced) measures of position, not absolute measures of content mastery.
• Example: A pupil with a percentile rank of 85 in the National Achievement Test performed as well as or better than 85% of all pupils who took the test — it does NOT mean the pupil answered 85% of items correctly.

Definitions

Section Title

Quartiles, Deciles, and Percentiles

Common Mistakes

• Treating percentile rank as a percentage-correct score — a PR of 90 means outperforming 90% of the group, NOT getting 90% of items right.
• Forgetting that Q2 = P50 = D5 = the median.
• Confusing quartiles (4 groups) with deciles (10 groups) with percentiles (100 groups).
• Saying Q1 = 25th score instead of the 25th PERCENTILE — quartiles are points, not score positions.

Formulas

Exam Tips

• If an LET item gives you mean and SD and asks for the score range covering 95% of pupils, apply Mean ± 2SD.
• Normal curve = symmetrical = Mean = Median = Mode. Any item that breaks this symmetry is pointing toward skewness.
• Know that under the normal curve, 50% of scores fall above the mean and 50% below — the mean is also the median.

Key Points

• The normal (bell) curve is a symmetrical, unimodal distribution that many large sets of human test scores approximate.
• Key property: MEAN = MEDIAN = MODE — all three measures of central tendency coincide at the center.
• The curve is asymptotic — the tails approach but never touch the baseline.
• The 68-95-99.7 Rule (Empirical Rule): approximately 68% of scores fall within 1 SD of the mean, 95% within 2 SDs, and 99.7% within 3 SDs.
• Example: If mean = 80 and SD = 5, then: 68% of pupils scored between 75 and 85 (80 ± 5); 95% scored between 70 and 90 (80 ± 10); 99.7% scored between 65 and 95 (80 ± 15).
• The normal curve is the theoretical basis for z-scores, T-scores, stanines, and standard score interpretations.
• The curve is perfectly symmetrical — 50% of scores fall above the mean and 50% below.

Definitions

Section Title

The Normal Curve

Common Mistakes

• Forgetting that mean = median = mode ONLY in a normal distribution — in skewed distributions they differ.
• Applying the 68-95-99.7 rule to non-normal (heavily skewed) distributions.
• Confusing the normal curve with any bell-shaped curve — a distribution must be symmetric to be truly normal.
• Thinking the tails of the normal curve eventually touch the x-axis (asymptotic property).

Formulas

Exam Tips

• FASTEST LOGIC: Where are MOST scores? If most scores are LOW → positive skew → difficult test. If most scores are HIGH → negative skew → easy test.
• Check your answer with the mean-median-mode order: if you said positive skew, confirm Mean > Median > Mode.
• LET scenario items often describe classroom performance (e.g., 'majority of pupils obtained high marks'). Map this directly: high marks → negative skew → easy test.
• Never memorize the skewness direction without reasoning it through — the tail follows the outliers (few extreme high or few extreme low scores).

Key Points

• Skewness describes the ASYMMETRY of a distribution. A skewed distribution has a longer tail on one side.
• CRITICAL RULE: Skewness is named after the TAIL, not the hump (where most scores are).
• POSITIVELY SKEWED (skewed to the RIGHT): The tail points toward the HIGH end. Most scores are LOW. The test was DIFFICULT (or the class performed poorly). Order of measures: Mean > Median > Mode.
• NEGATIVELY SKEWED (skewed to the LEFT): The tail points toward the LOW end. Most scores are HIGH. The test was EASY (or the class performed well). Order of measures: Mean < Median < Mode.
• The MEAN is dragged toward the tail (it chases outliers). This is why the mean sits above the median in positive skew and below it in negative skew.
• SCENARIO ITEM LOGIC: If most pupils got HIGH scores with only a few getting low → distribution is NEGATIVELY SKEWED → test was EASY.
• SCENARIO ITEM LOGIC: If most pupils got LOW scores with only a few getting high → distribution is POSITIVELY SKEWED → test was DIFFICULT.
• Memory aid: POSitive skew = POor performance (low scores dominate); NEGative skew = goNE well (high scores dominate).

Definitions

Section Title

Skewness — The Classic LET Trap

Common Mistakes

• Naming the skew after the HUMP instead of the TAIL — most students make this error. If most scores are low and the tail goes right, it is POSITIVE skew, not negative.
• Reversing the order of Mean, Median, and Mode for positive vs. negative skew.
• Confusing 'difficult test' with 'positive skew' direction — remember: difficulty pushes scores DOWN (left hump) and the tail stretches UP (right = positive).
• Saying the mean equals the median in a skewed distribution — they only coincide in a normal (symmetric) distribution.
• Forgetting that in ANY skewed distribution, the MEAN is the least representative measure — always use the median for skewed data.

Formulas

Exam Tips

• Classic LET item: Given two subjects with different means and SDs, compute z for each and compare. The subject with the HIGHER z-score is the one with better relative performance.
• Memorize both formulas: z = (X − x̄)/SD and T = 50 + 10z. Expect at least one computation item on each.
• z-score of 0 means the student scored EXACTLY at the mean — neither above nor below average.
• If a T-score is given and you need z: rearrange to z = (T − 50)/10.
• Under the normal curve: z = +1 corresponds to about the 84th percentile; z = −1 to the 16th percentile.

Key Points

• Raw scores from different tests cannot be compared directly. A score of 40 on a 50-item quiz and a score of 40 on an 80-item exam have different meanings.
• Standard scores solve this by expressing raw scores in STANDARD DEVIATION units, placing them on a common scale.
• The Z-SCORE tells how many SDs a score lies ABOVE (+) or BELOW (−) the mean. The z-distribution has a mean of 0 and an SD of 1.
• A positive z-score means the pupil scored ABOVE the class average; a negative z-score means BELOW.
• The T-SCORE converts z-scores to a scale with a mean of 50 and an SD of 10 — no negative scores in practice, easier to explain to parents.
• To compare a student's performance ACROSS subjects, compute the z-score for each subject — the higher z-score indicates better RELATIVE performance.
• Worked Example — Carlo: Math score 85, class mean 80, SD 5; English score 88, class mean 86, SD 4. Math z = (85−80)/5 = +1.0. English z = (88−86)/4 = +0.5. Carlo performed BETTER in Math despite the lower raw score.
• Stanines, NCE scores, and scaled scores are all derived from z-scores — knowing the z-score formula anchors all standard score conversions.

Definitions

Section Title

Standard Scores: z-score and T-score

Common Mistakes

• Thinking a higher RAW score always means better relative performance — always use z-scores to compare across subjects.
• Forgetting the sign of the z-score — positive means above the mean, negative means below.
• Applying the T-score formula incorrectly: T = 50 + 10z, NOT T = 50 + z/10.
• Computing z as SD/(X − mean) instead of (X − mean)/SD — the formula is deviation divided by SD, not the reverse.
• Assuming z-scores require a normal distribution to compute — z-scores can be computed for any distribution, but their percentile interpretation requires normality.

Formulas

Exam Tips

• Strength comparison: always compare ABSOLUTE values. The number closest to 1 (regardless of sign) is the strongest correlation.
• If an LET item asks for the correlation between two reliability estimates, expect r values between 0 and +1.00 (well-designed tests should have high positive reliability coefficients).
• Correlation ≠ causation — if an item offers a causal explanation based solely on correlation data, that option is INCORRECT.

Key Points

• A correlation coefficient (r) describes the RELATIONSHIP between two sets of scores or variables.
• r carries TWO pieces of information: DIRECTION (positive or negative) and STRENGTH (magnitude from 0 to 1).
• POSITIVE CORRELATION: Both variables move in the SAME direction (e.g., more study hours → higher grades).
• NEGATIVE CORRELATION: Variables move in OPPOSITE directions (e.g., more absences → lower grades).
• r ranges from −1.00 (perfect negative) through 0.00 (no relationship) to +1.00 (perfect positive).
• STRENGTH depends on the ABSOLUTE VALUE, not the sign: r = −0.85 is STRONGER than r = +0.60.
• Strength benchmarks: 0.00–0.20 = negligible; 0.21–0.40 = low; 0.41–0.60 = moderate; 0.61–0.80 = high; 0.81–1.00 = very high.
• CORRELATION DOES NOT IMPLY CAUSATION — this is a critical conceptual principle.
• Reliability coefficients (test-retest, parallel forms) and criterion-related validity coefficients are all correlation coefficients.
• The Pearson r is used for interval/ratio data; the Spearman rho is used for ordinal/ranked data.

Definitions

Section Title

Correlation

Common Mistakes

• Confusing direction with strength — r = −0.85 is STRONGER than r = +0.60 because |-0.85| = 0.85 > 0.60.
• Concluding causation from correlation — two variables may correlate without one causing the other.
• Thinking r = 0 means no relationship at all — it means no LINEAR relationship; a curved relationship can still exist.
• Confusing Pearson r (interval data) with Spearman rho (ordinal/ranked data).

Formulas

Exam Tips

• LET computation: Memorize the three-step process: (1) PS = raw/highest × 100, (2) Weighted Score = PS × weight, (3) Initial Grade = sum of weighted scores, then transmute.
• Know the component weights by subject type: Languages/AP/EsP = 30-50-20; Science/Math = 40-40-20; MAPEH/EPP/TLE = 20-60-20. QA is ALWAYS 20.
• Transmutation fact: 60 → 75 (minimum passing) and 100 → 100. All points between are proportionally mapped.
• Grade descriptor for the passing grade (75–79) = FAIRLY SATISFACTORY. Students who get below 75 = DID NOT MEET EXPECTATIONS.
• Final grade = average of Q1+Q2+Q3+Q4. If one quarterly grade is given and you need to find another to reach a target final grade, use algebra.

Key Points

• DepEd Order No. 8, s. 2015 is the policy guideline on classroom assessment for the K to 12 Basic Education Program — STANDARDS-BASED and COMPETENCY-BASED.
• Every quarterly grade is computed from THREE COMPONENTS: (1) Written Work (WW), (2) Performance Tasks (PT), and (3) Quarterly Assessment (QA).
• WRITTEN WORK includes quizzes, long tests, unit tests, written essays, and other written outputs.
• PERFORMANCE TASKS include demonstrations, projects, experiments, presentations, performances, and portfolios — the heaviest weighted component in most subjects.
• QUARTERLY ASSESSMENT is the exam (or performance) given at the end of each quarter.
• COMPONENT WEIGHTS for Grades 1–10 vary by subject area. Language subjects (Filipino, English) and AP and EsP: WW 30%, PT 50%, QA 20%. Science and Math: WW 40%, PT 40%, QA 20%. MAPEH and EPP/TLE: WW 20%, PT 60%, QA 20%.
• Notice: QA is always 20% across ALL subject areas in Grades 1–10.
• COMPUTATION STEPS: (1) Convert raw scores to Percentage Score (PS) = raw/highest × 100. (2) Multiply PS by weight to get Weighted Score. (3) Add Weighted Scores = Initial Grade. (4) Apply TRANSMUTATION TABLE to get Quarterly Grade.
• TRANSMUTATION: Initial grade of 100 transmutes to 100; Initial grade of 60 transmutes to 75 (the minimum passing grade). All initial grades below 60 transmute proportionally down to a minimum reported grade of 60.
• GRADE DESCRIPTORS: 90–100 = Outstanding; 85–89 = Very Satisfactory; 80–84 = Satisfactory; 75–79 = Fairly Satisfactory; Below 75 = Did Not Meet Expectations.
• PASSING GRADE = 75. The LOWEST grade that appears on the Report Card = 60.
• FINAL GRADE = average of the four quarterly grades.
• Pupils who fail to meet the minimum required grade are given REMEDIATION opportunities. Retention and promotion rules follow the DepEd policy.

Definitions

Section Title

The DepEd K-12 Grading System (DepEd Order No. 8, s. 2015)

Common Mistakes

• Using the same component weights for all subjects — weights DIFFER by subject area; QA is always 20% but WW and PT vary.
• Confusing the initial grade with the final reported grade — the initial grade must be TRANSMUTED first.
• Thinking the lowest grade that can appear on a report card is 75 — it is actually 60 (below 75 = Did Not Meet Expectations but the card still shows down to 60).
• Forgetting that the final grade is the simple average of the FOUR quarterly grades.
• Mixing up Written Work with Quarterly Assessment — WW is ongoing (quizzes, unit tests throughout the quarter); QA is the single end-of-quarter exam.
• Applying Senior High School weights to Grades 1–10 — SHS has its own weight sets by core, applied, and specialized subjects.