Raven's Progressive Matrices Test Raven's Progressive Matrices Distribution and Frequency Rules 3 — Questions and Answers
Question 1: On the Advanced Progressive Matrices (APM), why is the score distribution typically more negatively skewed than on the Standard Progressive Matrices?
- APM has more items overall
- APM is designed for above-average ability groups, so most scores cluster near the top (Correct answer)
- APM uses a different scoring method
- APM norms were collected from a wider age range
Correct answer: APM is designed for above-average ability groups, so most scores cluster near the top
The APM targets high-ability individuals, so their scores bunch near the ceiling, creating negative skew with few low-end scores.
Question 2: In RPM frequency tables published in manuals, what does the cumulative frequency column represent?
- The number of people who scored exactly at that raw score
- The sum of all frequencies from the lowest score up to and including the current score (Correct answer)
- The percentage of people above the current score
- The average of all scores up to that point
Correct answer: The sum of all frequencies from the lowest score up to and including the current score
Cumulative frequency is the running total of all frequencies from the lowest score through the current score level.
Question 3: If an RPM norming sample has a mean of 42 and a standard deviation of 8, what raw score corresponds approximately to the 84th percentile?
- 34
- 42
- 50 (Correct answer)
- 58
Correct answer: 50
In a normal distribution, one standard deviation above the mean (42 + 8 = 50) corresponds to approximately the 84th percentile.
Question 4: What is 'ceiling effect' in the context of RPM score distributions?
- When many test-takers score at or near the maximum possible score, compressing the upper tail of the distribution (Correct answer)
- When older adults score higher than younger adults
- When the test has too many difficult items for most respondents
- When a test-taker refuses to attempt the final items
Correct answer: When many test-takers score at or near the maximum possible score, compressing the upper tail of the distribution
Ceiling effect occurs when scores cluster near the maximum, truncating the distribution's upper end and reducing its ability to differentiate high scorers.
Question 5: When RPM norms are expressed as T-scores, what are the mean and standard deviation of the T-score distribution?
- Mean = 100, SD = 15
- Mean = 50, SD = 10 (Correct answer)
- Mean = 0, SD = 1
- Mean = 10, SD = 3
Correct answer: Mean = 50, SD = 10
T-scores are standardized to a mean of 50 and standard deviation of 10, making them one common way to express RPM performance.
Question 6: Which feature of a frequency distribution would indicate that RPM scores in a sample are uniformly distributed?
- A bell-shaped curve with most scores near the center
- Approximately equal frequencies at every score level (Correct answer)
- A sharp peak at one score value with few others
- Frequencies increasing steadily from low to high scores
Correct answer: Approximately equal frequencies at every score level
A uniform distribution has roughly equal counts at each score, producing a flat frequency histogram with no single dominant score range.
Question 7: In RPM normative tables, 'percentile equivalent' and 'percentile rank' are sometimes distinguished: which statement is accurate?
- Percentile rank always equals the raw score divided by the maximum score times 100
- Both terms refer identically to the percentage of examinees scoring at or below a given value (Correct answer)
- Percentile equivalent refers only to age equivalents, not rank
- Percentile rank is used only in clinical settings while percentile equivalent is used in research
Correct answer: Both terms refer identically to the percentage of examinees scoring at or below a given value
In RPM manuals both 'percentile rank' and 'percentile equivalent' indicate the proportion of the norm group at or below a given raw score.
On the Advanced Progressive Matrices (APM), why is the score distribution typically more negatively skewed than on the Standard Progressive Matrices?