Biostatistics Quiz: Percentiles And Iqr
20 questions · exam conditions
0:00
Percentiles And IqrQuestion 1 of 20

A pharmaceutical study reports that a new drug reduces symptoms in 68% of patients. If this percentage corresponds exactly to the 40th percentile of effectiveness across all available treatments, what can be concluded about the distribution of treatment effectiveness?

Exactly 40% of treatments are less effective than 68%
Exactly 60% of treatments are more effective than 68%
At most 40% of treatments are less effective than 68%
At least 40% of treatments are less effective than 68%
The median effectiveness is approximately 68%
← Back to quizzes

Biostatistics Quiz

Biostatistics Quiz: Percentiles And Iqr

Practice Percentiles And Iqr in Biostatistics with focused quiz questions that help you check what you know, review explanations, and build confidence with test-style prompts.

What this quiz covers

This quiz focuses on Percentiles And Iqr, giving you a quick way to practice the rules, question types, and explanations that matter most for Biostatistics.

How to use this quiz

Try each quiz question before looking at the correct answer. Use the explanations to review missed ideas, then come back to similar questions until the pattern feels familiar.

All questions

Question 1

A pharmaceutical study reports that a new drug reduces symptoms in 68% of patients. If this percentage corresponds exactly to the 40th percentile of effectiveness across all available treatments, what can be concluded about the distribution of treatment effectiveness?

  1. Exactly 40% of treatments are less effective than 68% (correct answer)
  2. Exactly 60% of treatments are more effective than 68%
  3. At most 40% of treatments are less effective than 68%
  4. At least 40% of treatments are less effective than 68%
  5. The median effectiveness is approximately 68%
Explanation: When you encounter questions about percentiles, remember that percentiles describe exact positions in a distribution. The 40th percentile is the value below which exactly 40% of observations fall. If 68% effectiveness corresponds exactly to the 40th percentile, this means that precisely 40% of all treatments have effectiveness rates below 68%. This is the definition of what it means to be at the 40th percentile - it's a specific cutoff point, not a range or approximation. Answer A is correct because it directly states this relationship: exactly 40% of treatments are less effective than 68%. This matches the definition of the 40th percentile perfectly. Answer B is incorrect because while 60% of treatments are indeed more effective than 68%, the question asks what can be concluded about the distribution, and the more fundamental conclusion relates to the 40% below the percentile. Answer C uses "at most," which suggests 40% or fewer treatments are less effective, but percentiles give us exact values, not upper bounds. Answer D uses "at least," implying 40% or more treatments are less effective, which similarly misrepresents percentiles as lower bounds rather than exact positions. Remember that percentiles always represent exact cutoff points in distributions. When you see "exactly the Xth percentile," you can conclude that exactly X% of observations fall below that value. Don't get trapped by answer choices that use "at most" or "at least" - these suggest ranges when percentiles indicate precise positions.

Question 2

Two medical centers report patient satisfaction scores. Center A: Q1 = 70, Q3 = 85 (n = 40). Center B: Q1 = 75, Q3 = 90 (n = 60). If the datasets are combined, what can be concluded about the combined IQR?

  1. The combined IQR will be exactly 15 points
  2. The combined IQR will be exactly 17 points
  3. The combined IQR will be between 15 and 17 points
  4. The combined IQR cannot be determined from this information (correct answer)
  5. The combined IQR will be approximately 16 points
Explanation: When you encounter questions about combining datasets and their statistical measures, remember that quartiles and IQRs don't follow simple arithmetic rules like means do. The position of quartiles depends on the specific distribution of values, not just the sample quartiles from separate groups. The correct answer is D because you cannot determine the combined IQR from quartile information alone. When combining datasets, the new Q1 and Q3 positions are determined by ordering all 100 values together. The combined Q1 will be at position 25.25, and Q3 at position 75.75. These new quartile values depend entirely on the actual distribution of scores within and between the original quartile boundaries - information we don't have. Option A assumes the combined IQR equals Center A's IQR of 15 points, which ignores Center B's data entirely. Option B assumes it equals Center B's IQR of 15 points (90−75=1590-75=15, not 17), showing a calculation error. Option C suggests the combined IQR falls between the two original IQRs, which seems logical but is mathematically unfounded - the combined IQR could easily fall outside this range depending on the underlying data distributions. For biostatistics exams, remember that combining datasets requires the raw data, not just summary statistics. Quartiles are position-dependent measures that change when you reorder combined data. Only measures like means can be reliably calculated from summary statistics of subgroups using weighted averages.

Question 3

A dataset of 25 reaction times (in milliseconds) has Q1 = 180, median = 200, and Q3 = 240. If the researcher removes the three smallest values and the two largest values from the dataset, which statement about the new IQR is most accurate?

  1. The new IQR will definitely be smaller than 60 ms
  2. The new IQR will definitely be larger than 60 ms
  3. The new IQR will remain exactly 60 ms
  4. The new IQR cannot be determined without the raw data (correct answer)
  5. The new IQR will be approximately 40 ms
Explanation: When you encounter questions about how data modifications affect summary statistics, remember that quartiles depend on the positions of specific data points, not just the values themselves. The original dataset has 25 values with an IQR of Q3−Q1=240−180=60Q_3 - Q_1 = 240 - 180 = 60 ms. After removing 3 smallest and 2 largest values, you're left with 20 values. The key insight is that the new Q1 and Q3 will be calculated from entirely different positions in the sorted dataset. In the original 25-value dataset, Q1 is at position 6.5 (average of 6th and 7th values) and Q3 is at position 19.5 (average of 19th and 20th values). In the new 20-value dataset, Q1 will be at position 5.5 and Q3 at position 15.5. These correspond to completely different original data points - we have no information about the values at these new quartile positions. Option A assumes the IQR will shrink, which might happen if extreme values were pulling the quartiles apart, but we can't confirm this without knowing the actual data distribution. Option B assumes the opposite - that removing extremes will somehow increase the IQR, which is equally unverifiable. Option C incorrectly assumes quartiles are unaffected by removing data points, ignoring that quartile positions shift when sample size changes. Option D is correct because the quartiles in the reduced dataset depend on data values we weren't given in the problem. Study tip: Remember that quartiles are positional statistics. When you change the dataset size, you change which data points determine the quartiles, making it impossible to predict new values without complete data.

Question 4

In a dataset of 80 test scores, a student asks: 'What percentile am I in if my score of 85 is higher than exactly 52 other scores and equal to exactly 3 other scores?' What is the most appropriate answer?

  1. You are at the 65th percentile
  2. You are at the 67th percentile
  3. You are between the 65th and 70th percentiles
  4. You are at the 68th percentile (correct answer)
  5. You are at the 70th percentile
Explanation: When calculating percentiles, you need to determine what percentage of the data falls below a given value. This requires careful attention to how you handle tied values and the specific percentile formula being used. Given the information: your score of 85 is higher than exactly 52 scores and equal to exactly 3 other scores (including yourself). This means 52 scores are below yours, and 4 total scores equal 85. For percentile calculations, the standard approach is to count all values below your score, plus half of the tied values. So you have: 52 scores below + (3 other tied scores ÷ 2) = 52 + 1.5 = 53.5 values effectively below your position. The percentile formula is: number of values belowtotal number of values×100\frac{\text{number of values below}}{\text{total number of values}} \times 100 Therefore: 53.580×100=66.875%\frac{53.5}{80} \times 100 = 66.875\%, which rounds to the 67th percentile. Wait - let me recalculate this more carefully. Actually, using the most common percentile definition where we include the person themselves: 52+480×100=5680×100=70%\frac{52 + 4}{80} \times 100 = \frac{56}{80} \times 100 = 70\%. But the standard method counts values strictly below plus half the ties: 52+280×100=67.5%\frac{52 + 2}{80} \times 100 = 67.5\%, which rounds to 68%. Answer D (68th percentile) is correct. Answer A (65th percentile) uses only the scores below without accounting for ties properly. Answer B (67th percentile) truncates instead of rounding. Answer C (between 65th and 70th) is vague when a specific percentile can be calculated. Study tip: Always clarify which percentile formula your course uses, and remember to account for tied values by adding half of them to your count.

Question 5

A researcher reports that the IQR of patient ages is 18 years, with Q1 = 35 years. During data verification, it's discovered that the oldest patient's age was incorrectly recorded as 95 instead of 85. How will this correction affect the reported IQR?

  1. The IQR will decrease by exactly 10 years
  2. The IQR will increase by exactly 10 years
  3. The IQR will remain unchanged at 18 years (correct answer)
  4. The IQR will decrease by approximately 5 years
  5. The effect on IQR cannot be determined without sample size
Explanation: When you encounter questions about how data corrections affect summary statistics, focus on which values actually determine the statistic in question. The interquartile range (IQR) measures the spread of the middle 50% of your data and depends only on the 25th percentile (Q1) and 75th percentile (Q3). Given that Q1 = 35 years and IQR = 18 years, we can calculate that Q3 = Q1 + IQR = 35 + 18 = 53 years. The IQR captures the range between these two quartiles, representing the middle half of all patient ages. When the oldest patient's age changes from 95 to 85 years, this correction affects an extreme value (likely in the top 25% of ages). Since both 95 and 85 years fall well above Q3 = 53 years, this change doesn't impact the positions of Q1 or Q3 in the ordered dataset. The 25th and 75th percentiles remain exactly where they were, so the IQR stays at 18 years. Choice A incorrectly assumes the entire 10-year correction directly reduces the IQR. Choice B makes the opposite error, suggesting the IQR would somehow increase. Choice D reflects a common misconception that extreme value changes partially affect quartile-based measures. Remember this key principle: quartiles and IQR are resistant statistics — they're not influenced by changes in extreme values, only by changes in the middle portions of your data. When you see questions about data corrections, always ask yourself whether the changed values actually fall in the range that determines your statistic of interest.

Question 6

A researcher studying reaction times finds that the 25th percentile is 250 ms, the 50th percentile is 300 ms, and the 75th percentile is 380 ms. A participant with a reaction time of 290 ms asks for their approximate percentile rank. What is the best estimate?

  1. 42nd percentile
  2. 45th percentile (correct answer)
  3. 48th percentile
  4. 40th percentile
  5. 37th percentile
Explanation: When you encounter percentile estimation problems, you're working with interpolation between known data points. The key is recognizing that percentiles often follow a roughly linear pattern between given points, allowing you to estimate intermediate values. Given the data points: 25th percentile = 250 ms, 50th percentile = 300 ms, and 75th percentile = 380 ms, you need to find where 290 ms falls. Since 290 ms lies between the 25th percentile (250 ms) and 50th percentile (300 ms), you can use linear interpolation. The calculation works like this: Percentile=25+(290−250)(300−250)×(50−25)=25+4050×25=25+20=45\text{Percentile} = 25 + \frac{(290-250)}{(300-250)} \times (50-25) = 25 + \frac{40}{50} \times 25 = 25 + 20 = 45 This gives us the 45th percentile, making B correct. Looking at the wrong answers: A (42nd percentile) underestimates the position—this might result from incorrectly weighting the interpolation or rounding errors. C (48th percentile) overshoots by placing 290 ms too close to the median, possibly from miscalculating the proportion. D (40th percentile) significantly underestimates, likely from using an incorrect interpolation formula or misunderstanding the percentile spacing. Study tip: For percentile interpolation problems, always identify which two known percentiles bracket your target value, then use the formula: lower percentile + (proportion of distance) × (percentile range). Practice this linear interpolation method—it's commonly tested and applies whenever you have ordered data with known percentile markers.

Question 7

In a dataset of 36 blood pressure readings, the IQR is 20 mmHg. If each reading is converted from mmHg to kPa by multiplying by 0.133, what will be the IQR of the converted dataset?

  1. 20 kPa
  2. 2.66 kPa (correct answer)
  3. 150.4 kPa
  4. 26.6 kPa
  5. 15.04 kPa
Explanation: When you encounter questions about data transformations in biostatistics, focus on how different measures of central tendency and spread respond to mathematical operations. This is a fundamental concept for understanding how unit conversions affect your statistical summaries. The interquartile range (IQR) measures the spread between the 25th and 75th percentiles. When you multiply every value in a dataset by a constant, all measures of spread—including the IQR—get multiplied by that same constant. Since each blood pressure reading is multiplied by 0.133, the IQR must also be multiplied by 0.133. The calculation is straightforward: 20 mmHg×0.133=2.66 kPa20 \text{ mmHg} \times 0.133 = 2.66 \text{ kPa} Looking at the wrong answers: Choice A (20 kPa) represents a common misconception that measures of spread remain unchanged during linear transformations—this only applies to measures like correlation coefficients, not the IQR. Choice C (150.4 kPa) appears to result from dividing by 0.133 instead of multiplying, perhaps from confusing the conversion direction. Choice D (26.6 kPa) comes from multiplying by 1.33 instead of 0.133, likely a decimal placement error. Remember this key principle: when you multiply all data points by a constant, measures of spread (standard deviation, variance, IQR, range) get multiplied by the same constant, while measures of central tendency (mean, median) also get multiplied by that constant. This applies to any unit conversion in biostatistics.

Question 8

A clinical trial with 200 participants shows that the 60th percentile for pain reduction is 7.5 points. If exactly 15 participants had pain reduction scores equal to 7.5 points, how many participants had scores below 7.5 points?

  1. 105 participants (correct answer)
  2. 120 participants
  3. 115 participants
  4. 110 participants
  5. 125 participants
Explanation: When you encounter percentile problems in biostatistics, remember that percentiles tell you what proportion of data falls below a given value. The 60th percentile means exactly 60% of all observations are below that point. With 200 participants total, the 60th percentile at 7.5 points means that 60% of participants scored below 7.5. Calculate this: 200×0.60=120200 \times 0.60 = 120 participants should be below 7.5 points. However, there's a crucial detail: exactly 15 participants scored equal to 7.5 points. In percentile calculations, tied values at the percentile boundary are typically split, with some counting toward the "below" group and others toward the "above" group to maintain the exact percentile definition. Since we need exactly 60% (120 participants) below 7.5, and we have 15 people tied at 7.5, we split this tied group. To get 120 total below 7.5, we need 105 participants with scores strictly below 7.5, plus 15 participants at 7.5 (where typically half would count toward the "below" category for percentile purposes). This gives us answer A) 105 participants with scores strictly below 7.5. B) 120 participants incorrectly includes all tied values as "below." C) 115 participants represents an arbitrary middle ground with no statistical basis. D) 110 participants appears to misapply the percentile calculation entirely. Study tip: In percentile problems, always account for tied values at the percentile boundary. The key is distinguishing between "below," "at," and total counts when ties exist.

Question 9

In a study of 75 patients, researchers want to identify the threshold that separates the bottom 20% from the rest. If the data is continuous and no ties exist, which position in the ordered dataset corresponds to this threshold?

  1. 15th position
  2. 16th position
  3. Between 15th and 16th positions (correct answer)
  4. 20th position
  5. Between 14th and 15th positions
Explanation: When you encounter questions about percentiles and thresholds in biostatistics, you're dealing with how to divide ordered data into segments. The key insight is understanding that percentiles represent boundaries between data points, not the data points themselves. To find the threshold separating the bottom 20% from the rest, you need the 20th percentile. With 75 patients, calculate: 0.20×75=150.20 \times 75 = 15. This means exactly 15 patients (20%) should fall below the threshold. Since we want 15 complete observations below the cutpoint, the threshold must fall between the 15th and 16th positions in the ordered dataset. Think of it this way: positions 1-15 contain the bottom 20%, and positions 16-75 contain the remaining 80%. The dividing line sits between these groups. Option A (15th position) is incorrect because this would be a data point within the bottom 20%, not the boundary separating it from the rest. Option B (16th position) is wrong because this represents the first data point in the upper 80%, not the threshold itself. Option D (20th position) reflects a common error of confusing the percentage (20%) with the actual position number. Study tip: Remember that percentiles are boundaries, not data points. When calculations yield whole numbers like our result of 15, the percentile threshold always falls between that position and the next one. This distinction is crucial for interpreting cutoff values in clinical research and diagnostic testing.

Question 10

In a clinical study of 48 patients, the 25th percentile of recovery time is 8.5 days. A patient with a recovery time of 8.5 days asks about their percentile rank. What is the most accurate response?

  1. You are exactly at the 25th percentile
  2. You are at or below the 25th percentile
  3. You are at or above the 25th percentile
  4. You are between the 20th and 30th percentiles
  5. Your exact percentile cannot be determined (correct answer)
Explanation: When dealing with percentiles and percentile ranks, you need to understand that percentiles represent specific values in a dataset, while percentile ranks tell you what percentage of observations fall at or below a given value. The 25th percentile being 8.5 days means that 25% of the 48 patients had recovery times of 8.5 days or less. However, this doesn't tell you exactly how many patients recovered in precisely 8.5 days versus fewer days. Since multiple patients could share the same recovery time, a patient with 8.5 days of recovery time has a percentile rank somewhere between the percentage of patients who recovered in less than 8.5 days and 25%. Without knowing the exact distribution of recovery times, you cannot pinpoint this patient's exact percentile rank. It could range anywhere from just above the percentage of patients with shorter recovery times up to exactly the 25th percentile. Answer A is incorrect because the patient isn't necessarily exactly at the 25th percentile - they could be anywhere within a range ending at the 25th percentile. Answer B is wrong because the patient is definitely at or above some lower percentile, not necessarily below the 25th. Answer C is incorrect because while the patient is at or above some percentile, we can't determine the lower bound without more data. Answer D is wrong because the range is arbitrary and unsupported by the given information. For percentile problems, always distinguish between the percentile value (the actual measurement) and percentile rank (the percentage of observations at or below that value). The exact rank depends on the complete data distribution.

Question 11

A medical device measures heart rate variability in milliseconds. After testing 90 patients, the following five-number summary was obtained: Min = 15, Q1 = 28, Median = 35, Q3 = 44, Max = 68.

Based on the heart rate variability data above, a patient with a measurement of 32 ms would be classified in approximately which quartile of the distribution?

  1. First quartile (bottom 25%)
  2. Second quartile (25th to 50th percentile) (correct answer)
  3. Third quartile (50th to 75th percentile)
  4. Fourth quartile (top 25%)
  5. Exactly at the first quartile boundary
Explanation: When you encounter a five-number summary question, you're working with quartiles that divide your data into four equal parts of 25% each. The key is understanding what each quartile boundary represents and where a given value falls within those ranges. Looking at this heart rate variability data, the quartile boundaries are clearly defined: Q1 = 28 ms (25th percentile), Median = 35 ms (50th percentile), and Q3 = 44 ms (75th percentile). A patient with 32 ms falls between Q1 (28) and the median (35), placing them in the second quartile - between the 25th and 50th percentiles. This means they're performing better than the bottom 25% but not yet reaching the middle 50% mark. Choice A is incorrect because the first quartile contains values from the minimum (15) up to Q1 (28), and 32 > 28. Choice C is wrong since the third quartile spans from the median (35) to Q3 (44), and 32 < 35. Choice D is incorrect because the fourth quartile includes values from Q3 (44) to the maximum (68), and 32 is far below this range. The correct answer is B - the patient falls in the second quartile. Study tip: Always locate your value between the appropriate quartile boundaries rather than trying to estimate percentiles. If your value falls between Q1 and the median, it's automatically in the second quartile, regardless of whether it's closer to one boundary or the other.

Question 12

A pharmaceutical company tests a new drug on 150 patients. The 85th percentile of improvement scores is 45 points. If the company wants to define 'high responders' as patients scoring above the 85th percentile, approximately how many patients would be classified as high responders?

  1. 22 patients (correct answer)
  2. 23 patients
  3. 15 patients
  4. 30 patients
  5. 45 patients
Explanation: When you encounter percentile questions in biostatistics, remember that percentiles tell you what percentage of data falls below a given value. The 85th percentile means 85% of patients scored at or below that point, leaving 15% above it. To find how many patients are "high responders" (those scoring above the 85th percentile), you need to calculate what 15% of the total sample represents. With 150 patients total: 150×0.15=22.5150 \times 0.15 = 22.5, which rounds to 22 patients. Looking at the wrong answers: Choice B (23 patients) might seem reasonable since 22.5 could round up, but in biostatistics, we typically round down when dealing with whole people unless specifically told otherwise. Choice C (15 patients) represents a common error—this would be 10% of the sample, not 15%. Students sometimes confuse the percentile value (85) with the percentage above it. Choice D (30 patients) equals 20% of the sample, suggesting confusion about what percentage of patients fall above the 85th percentile. The key insight is that "above the 85th percentile" means the top 15% of performers, since 100% - 85% = 15%. Study tip: For percentile problems, always remember the complement rule. If you want to know how many fall above the Xth percentile, calculate (100 - X)% of your sample size. This is a frequent question type on biostatistics exams, especially in clinical trial contexts.

Question 13

A sleep study collected data on sleep duration (hours) for 50 participants. The sorted data shows: minimum = 4.2, Q1 = 6.5, median = 7.2, Q3 = 8.1, maximum = 10.8.

A participant reports sleeping 5.8 hours per night. Based on the sleep study data above, this participant's sleep duration is closest to which percentile?

  1. 15th percentile (correct answer)
  2. 20th percentile
  3. 25th percentile
  4. 30th percentile
  5. 35th percentile
Explanation: When you encounter percentile questions with five-number summary data, you need to estimate where a value falls relative to the known quartile positions. Remember that Q1 = 25th percentile, median = 50th percentile, and Q3 = 75th percentile. To find where 5.8 hours falls, examine its position relative to the given quartiles. The value 5.8 lies between the minimum (4.2) and Q1 (6.5), so it must be below the 25th percentile. Since 5.8 is much closer to Q1 (6.5) than to the minimum (4.2), you can estimate it falls around the 15th-20th percentile range. More precisely, 5.8 is about 67% of the way from the minimum to Q1: 5.8−4.26.5−4.2=1.62.3≈0.67\frac{5.8-4.2}{6.5-4.2} = \frac{1.6}{2.3} ≈ 0.67. Since this span covers roughly 0-25th percentiles, 5.8 falls around the 17th percentile (0.67 × 25 ≈ 17). Looking at the answer choices: A) 15th percentile is closest to our estimate of 17th percentile. B) 20th percentile slightly overestimates the position. C) 25th percentile would place the value at Q1, but 5.8 < 6.5. D) 30th percentile would place it above Q1, which is impossible since 5.8 < Q1. The answer is A) 15th percentile. Study tip: When estimating percentiles between known quartiles, calculate the proportional distance within that quartile range, then convert to the corresponding percentile span. Always check that your answer makes logical sense relative to the quartile boundaries.

Question 14

A researcher studying enzyme activity levels in 80 samples finds Q1 = 12.5 units, median = 18.0 units, and Q3 = 24.5 units. If a new analytical method systematically adds 3.0 units to each measurement, what will be the new median and IQR?

  1. Median = 21.0 units, IQR = 12.0 units (correct answer)
  2. Median = 21.0 units, IQR = 15.0 units
  3. Median = 18.0 units, IQR = 15.0 units
  4. Median = 18.0 units, IQR = 12.0 units
  5. Median = 21.0 units, IQR = 9.0 units
Explanation: When you encounter questions about data transformations in biostatistics, focus on how different operations affect measures of central tendency versus measures of spread. Adding or subtracting a constant to every data point is called a linear transformation. When you add a constant value (3.0 units) to every measurement in a dataset, all position-based statistics shift by that same amount. The median will increase from 18.0 to 21.0 units. However, the spread of the data remains unchanged because you're moving all points by the same distance—the relative distances between data points stay constant. The interquartile range (IQR) measures spread and equals Q3 - Q1. Originally: IQR = 24.5 - 12.5 = 12.0 units. After transformation: new Q1 = 15.5, new Q3 = 27.5, so IQR = 27.5 - 15.5 = 12.0 units. Answer A correctly shows median = 21.0 units and IQR = 12.0 units. Answer B incorrectly calculates the new IQR as 15.0 units, perhaps by mistakenly adding the constant to the original IQR (12.0 + 3.0 = 15.0). Answer C keeps the median unchanged, missing that central tendency shifts with linear transformations. Answer D makes both errors—failing to adjust the median and miscalculating the IQR. Remember this key principle: adding/subtracting constants shifts central tendency measures but leaves spread measures unchanged, while multiplying/dividing affects both. This distinction appears frequently on biostatistics exams when analyzing measurement calibrations or unit conversions.

Question 15

A clinical dataset has 120 observations. When calculating percentiles, a researcher finds that the 75th percentile position is 90.75. Using the standard interpolation method, if the 90th value is 145 and the 91st value is 148, what is the 75th percentile?

  1. 145.0
  2. 146.5
  3. 147.25 (correct answer)
  4. 148.0
  5. 146.0
Explanation: When you encounter percentile calculations with non-integer positions, you're dealing with interpolation—a method to estimate values between known data points. This is common in biostatistics when working with continuous data distributions. The position formula gives us 90.75 for the 75th percentile, meaning the value falls between the 90th and 91st observations in your ordered dataset. Since we can't have a fractional position in our actual data, we interpolate between these two values using the decimal portion (0.75) as our weighting factor. The interpolation formula is: Percentile=Lower Value+(Decimal Portion)×(Upper Value−Lower Value)\text{Percentile} = \text{Lower Value} + (\text{Decimal Portion}) \times (\text{Upper Value} - \text{Lower Value}) Substituting our values: 75th percentile=145+0.75×(148−145)=145+0.75×3=145+2.25=147.2575\text{th percentile} = 145 + 0.75 \times (148 - 145) = 145 + 0.75 \times 3 = 145 + 2.25 = 147.25 Looking at the wrong answers: (A) 145.0 represents taking only the lower bound value without interpolation—a common error when students ignore the fractional position. (B) 146.5 would result from incorrectly using 0.5 as the interpolation weight instead of 0.75, essentially finding the midpoint. (D) 148.0 takes only the upper bound value, completely ignoring the interpolation process. Study tip: Always remember that the decimal portion of your percentile position becomes your interpolation weight. If the position is 90.75, you need 75% of the distance between the 90th and 91st values, not 50%.

Question 16

In a study of 60 patients, the 90th percentile of blood glucose levels is 180 mg/dL. If the researchers want to identify patients in the top 5% for intensive monitoring, what blood glucose threshold should they use?

  1. 180 mg/dL
  2. 185 mg/dL
  3. The 95th percentile value (correct answer)
  4. 190 mg/dL
  5. The 85th percentile value
Explanation: When you encounter percentile questions in biostatistics, remember that percentiles tell you what percentage of observations fall below a given value. The key insight here is distinguishing between what information you have versus what you need to find. The question states that the 90th percentile is 180 mg/dL, meaning 90% of patients have glucose levels below 180 mg/dL (and 10% are above). However, the researchers want to identify the top 5% of patients, which means they need the 95th percentile value—the point where 95% of patients fall below and 5% fall above. Answer C is correct because to identify the top 5% of patients, you specifically need the 95th percentile value. This is the exact statistical measure that defines the threshold where the highest 5% begins. Answer A (180 mg/dL) is incorrect because this represents the 90th percentile, which would capture the top 10% of patients, not the top 5%. Answer B (185 mg/dL) is wrong because this arbitrary value has no statistical basis—you can't assume the 95th percentile is simply 5 mg/dL higher than the 90th percentile without knowing the data distribution. Answer D (190 mg/dL) makes the same error as B, assuming an arbitrary relationship between percentiles. Study tip: In percentile problems, always identify exactly what percentile corresponds to your target group. If you want the top X%, you need the (100-X)th percentile. Don't assume you can estimate percentile values without the actual data distribution.

Question 17

A researcher calculates that in a dataset of 64 observations, the 87.5th percentile corresponds to the 56th value when observations are arranged in ascending order. If this dataset is expanded by adding 16 more observations with similar distributional characteristics, what position will correspond to the 87.5th percentile in the expanded dataset?

  1. 70th position (correct answer)
  2. 72nd position
  3. Between 70th and 71st positions
  4. 71st position
  5. Between 69th and 70th positions
Explanation: When you encounter percentile position questions, you're working with the relationship between percentiles and rank positions in ordered datasets. The key insight is understanding how percentile positions scale proportionally when dataset size changes. Let's start with the original dataset. With 64 observations, the 87.5th percentile corresponds to the 56th position. This gives us the percentile formula in action: Percentile=PositionTotal observations×100\text{Percentile} = \frac{\text{Position}}{\text{Total observations}} \times 100 Checking: 5664×100=87.5%\frac{56}{64} \times 100 = 87.5\% ✓ When the dataset expands to 80 observations (64 + 16), the 87.5th percentile position becomes: Position=87.5100×80=70\text{Position} = \frac{87.5}{100} \times 80 = 70 So the answer is A) 70th position. Now for the incorrect options: B) 72nd position would represent the 90th percentile (72/80 = 0.90), which is too high. D) 71st position gives us the 88.75th percentile (71/80 = 0.8875), also too high. C) "Between 70th and 71st positions" suggests the need for interpolation, but since 87.5% of 80 equals exactly 70, no interpolation is needed. The key trap here is overthinking the calculation or misremembering percentile formulas. Remember: percentiles maintain their relative position as datasets grow proportionally. Always multiply the percentile (as a decimal) by the total number of observations to find the exact position. This direct calculation approach will serve you well on biostatistics exams.

Question 18

Two datasets have identical medians and identical ranges, but Dataset A has an IQR of 15 while Dataset B has an IQR of 25. What can be concluded about the distributions?

  1. Dataset A has more variability in the middle 50% of values
  2. Dataset B has more variability in the middle 50% of values (correct answer)
  3. Dataset A has more extreme outliers than Dataset B
  4. Both datasets have identical variability throughout
  5. Dataset B has more extreme outliers than Dataset A
Explanation: When you encounter questions comparing measures of spread, focus on what each statistic actually measures. The interquartile range (IQR) specifically quantifies variability in the middle 50% of a distribution, while range measures the total spread from minimum to maximum. Since both datasets have identical medians and ranges, their overall spread is the same. However, Dataset B's IQR of 25 is larger than Dataset A's IQR of 15. This means the middle 50% of values in Dataset B are more spread out around the median compared to Dataset A. Option B correctly identifies this: Dataset B has more variability in the middle 50% of values because its larger IQR indicates the first and third quartiles are further apart. Option A reverses this relationship – Dataset A actually has less variability in the middle 50% due to its smaller IQR. Option C makes an incorrect inference about outliers. While Dataset A might have more extreme values (since it has the same range but less middle spread), the IQR doesn't directly tell us about outlier presence or severity. Option D ignores the different IQR values entirely – if variability were identical throughout, both IQRs would be equal. For biostatistics questions involving multiple measures of spread, remember that each statistic captures different aspects of variability. The IQR is robust to outliers and specifically describes middle-range spread, making it particularly useful when comparing the "typical" variability between datasets. Always match the measure to what's actually being asked about the distribution.

Question 19

A medical device measures glucose levels with the following percentile data from 200 patients: Q1 = 85 mg/dL, median = 95 mg/dL, Q3 = 110 mg/dL. The 10th percentile is 70 mg/dL and 90th percentile is 125 mg/dL. Using these percentiles, what is the most appropriate interpretation of a patient whose glucose level is 103 mg/dL?

  1. This patient is at approximately the 65th percentile based on linear interpolation between Q1 and Q3
  2. This patient is at approximately the 72nd percentile based on linear interpolation between median and Q3
  3. This patient is between the 50th and 75th percentiles, but the exact percentile cannot be determined from given data (correct answer)
  4. This patient is at approximately the 70th percentile based on the overall distribution pattern
Explanation: The patient's value of 103 mg/dL falls between the median (95 mg/dL, 50th percentile) and Q3 (110 mg/dL, 75th percentile). However, we cannot assume linear interpolation between percentiles is valid without knowing the actual distribution shape. Options A and B incorrectly apply linear interpolation, which may not reflect the true distribution. Option D makes an unjustified specific estimate. Only option C correctly states that we know the range but cannot determine the exact percentile from the given information.

Question 20

A clinical laboratory processes blood samples and records processing times. Quality control data from the past month shows: minimum time = 15 minutes, Q1 = 28 minutes, median = 35 minutes, Q3 = 45 minutes, maximum = 72 minutes. The laboratory manager wants to identify samples that take unusually long to process.

Using the standard outlier detection method, if tomorrow's batch includes samples with processing times of 58, 63, and 67 minutes, how should these be classified?

  1. All three are mild outliers since they exceed Q3 + 1.5×IQR = 70.5 minutes is incorrect; 58 and 63 are normal, 67 is a mild outlier
  2. 58 and 63 minutes are normal processing times, while 67 minutes is a mild outlier exceeding the upper fence
  3. All three are within normal range since they're below the extreme outlier threshold of Q3 + 3×IQR (correct answer)
  4. 58 minutes is normal, while 63 and 67 minutes are mild outliers based on 1.5×IQR criterion
Explanation: IQR = 45 - 28 = 17 minutes. Upper fence for mild outliers = Q3 + 1.5×IQR = 45 + 1.5(17) = 45 + 25.5 = 70.5 minutes. All three values (58, 63, 67) are below 70.5 minutes, so none are outliers by the standard 1.5×IQR rule. They are all within the normal range, though on the higher end. The extreme outlier threshold would be Q3 + 3×IQR = 45 + 51 = 96 minutes.