- Two datasets, X and Y, have the exact same mean, median, and mode. Which of the following statements must be true?
Answer: The two datasets do not necessarily have the same shape.
Measures of center (mean, median, mode) do not fully describe a distribution's shape or spread. It is possible to construct two different datasets that share the same mean, median, and mode but have different shapes and standard deviations. For example, Dataset 1: {1, 5, 5, 9} (Mean=5, Median=5, Mode=5) and Dataset 2: {4, 5, 5, 6} (Mean=5, Median=5, Mode=5). They share the same measures of center but have different spreads and shapes. Therefore, one cannot conclude they have the same shape.
- For a distribution that is moderately skewed to the left, which of the following best describes the typical relationship among the mean, median, and mode?
Answer: Mean < Median < Mode
In a skewed distribution, the mean is pulled in the direction of the long tail. For a left-skewed distribution, the tail is on the left, so the mean is pulled to the left (lower values). The mode is the peak of the distribution, which will be at a higher value. The median lies between the mean and the mode. Therefore, the typical ordering from least to greatest is mean, then median, then mode.
- Consider the following dataset of observations: 33, 35, 38, 38, 40, 41, 44, 45, 68. Which statement about this dataset is correct?
Answer: The value 68 is an outlier, and its presence makes the distribution right-skewed.
First, find the quartiles. With n=9, the median is the 5th value, which is 40. Q1 is the median of the lower half (33, 35, 38, 38), so Q1 = (35+38)/2 = 36.5. Q3 is the median of the upper half (41, 44, 45, 68), so Q3 = (44+45)/2 = 44.5. The IQR = 44.5 - 36.5 = 8. The upper fence is Q3 + 1.5*IQR = 44.5 + 1.5*8 = 44.5 + 12 = 56.5. Since 68 is greater than 56.5, it is an outlier. This high outlier pulls the mean to the right, making the distribution right-skewed.
- The histogram displays the distribution of a certain variable. Which of the following best describes the shape of the distribution?
Answer: Bimodal with peaks near 2 and 8
The histogram clearly shows two distinct peaks, or modes, in the data. One peak occurs in the bin centered around 2, and a second, slightly shorter peak occurs in the bin centered around 8. A distribution with two distinct peaks is called bimodal. It is not unimodal (one peak), nor is it uniform (all bars having roughly the same height).
- A researcher collects data on the response times of a cognitive task. Summary statistics for the dataset (n=50) are: Min=5.2, Q1=10.5, Median=12.1, Q3=14.2, Max=25.8. Using the 1.5 × IQR rule for identifying outliers, which of the following statements is true?
Answer: The maximum value of 25.8 is an outlier, but the minimum value of 5.2 is not.
First, calculate the Interquartile Range (IQR): IQR = Q3 - Q1 = 14.2 - 10.5 = 3.7. Next, calculate the lower and upper fences for outliers. Lower fence = Q1 - 1.5 × IQR = 10.5 - 1.5 × 3.7 = 10.5 - 5.55 = 4.95. Upper fence = Q3 + 1.5 × IQR = 14.2 + 1.5 × 3.7 = 14.2 + 5.55 = 19.75. The minimum value, 5.2, is greater than the lower fence of 4.95, so it is not an outlier. The maximum value, 25.8, is greater than the upper fence of 19.75, so it is an outlier.
- A strongly right-skewed distribution of data on household incomes is transformed by taking the natural logarithm of each income value. Which of the following is the most likely effect of this transformation on the shape of the distribution?
Answer: The distribution will become more symmetric.
Logarithmic transformations compress the scale of the data. For a right-skewed distribution, the high values in the right tail are very spread out. The log transformation has a larger effect on these large values than on the smaller values, effectively 'pulling in' the long right tail. This process tends to make the distribution more symmetric.
- Which of the following conditions, if true for a given dataset, is sufficient to conclude that the distribution is perfectly symmetric?
Answer: For every point above the median, there is a corresponding point an equal distance below the median.
This is the definition of symmetry around the median. If every point has a 'mirror image' on the other side of the median, the distribution is perfectly symmetric. The other conditions are necessary for many symmetric distributions but are not sufficient. For example, a distribution can have its mean equal to its median but still be slightly skewed if the tails are not balanced (A). The quartiles can be equidistant from the median, but the whiskers could have different lengths (B). The number of points above and below the mean can be equal in a non-symmetric distribution (D).
- A large dataset is approximately symmetric. The maximum value in the dataset is then erroneously replaced by a number ten times larger than the original maximum. How will this change affect the shape of the distribution and the interquartile range (IQR)?
Answer: The distribution will become right-skewed, and the IQR will likely remain unchanged.
Replacing the maximum value with a much larger number creates an extreme high outlier. This will pull the tail of the distribution to the right, making it right-skewed. The interquartile range (IQR = Q3 - Q1) is a resistant measure of spread because it is based on the middle 50% of the data. Since the change was made to the maximum value, it is very unlikely to affect either Q1 or Q3, especially in a large dataset. Therefore, the IQR will likely remain unchanged.
- For a particular dataset, a data point is found to have a z-score of 2.9. The distribution is known to be unimodal and only slightly skewed. Which of the following is a correct conclusion about this data point?
Answer: The point is unusual, but may or may not be classified as an outlier by the 1.5 × IQR rule.
A z-score of 2.9 indicates the data point is 2.9 standard deviations above the mean, which is an unusual value in any distribution. However, different rules for identifying outliers can lead to different conclusions. The common 'z-score > 3' rule is just one guideline. The 1.5 × IQR rule identifies outliers based on quartiles. For a perfectly normal distribution, the 1.5 × IQR fences correspond to z-scores of approximately ±2.7. Since this distribution is not specified as normal, we cannot be certain that a z-score of 2.9 falls outside the fences. Thus, the point is unusual, but its classification as an outlier by the IQR method is not guaranteed.
- A five-number summary for a set of exam scores is {20, 50, 60, 65, 100}. Based on the 1.5 × IQR rule, what can be concluded about outliers in this dataset?
Answer: There is at least one high outlier and at least one low outlier.
First, calculate the IQR: IQR = Q3 - Q1 = 65 - 50 = 15. Next, calculate the fences. The lower fence is Q1 - 1.5 × IQR = 50 - 1.5 × 15 = 50 - 22.5 = 27.5. The upper fence is Q3 + 1.5 × IQR = 65 + 1.5 × 15 = 65 + 22.5 = 87.5. The minimum value is 20, which is less than the lower fence of 27.5, so there is at least one low outlier. The maximum value is 100, which is greater than the upper fence of 87.5, so there is at least one high outlier.
- A financial analyst is studying the annual salaries of employees at a technology firm. The data is found to be roughly symmetric with a median salary of $95,000. The company then hires a new CEO with an annual salary of $15,000,000. How will the inclusion of the new CEO's salary most likely affect the mean and median salary, and the shape of the distribution?
Answer: The mean will increase substantially, the median will remain relatively stable, and the distribution will become right-skewed.
The CEO's salary is an extreme high outlier. The mean is sensitive to outliers and will be pulled upwards significantly by this large value. The median, being a resistant measure of center, will change very little, as it is based on the position of the middle value, not its magnitude. The addition of a high outlier creates a long tail on the right side of the distribution, making it right-skewed.
- A dataset contains an extreme, high outlier. If this outlier is identified and removed from the dataset, which of the following measures is expected to change the most?
Answer: The standard deviation
The standard deviation is calculated using the mean and measures the average distance of data points from the mean. Since the mean is heavily influenced by outliers, and the standard deviation calculation involves the squared differences from the mean, it is the most sensitive to extreme values. The median and IQR are resistant to outliers, and the mode is generally unaffected unless the outlier was part of a modal group.
- A dataset has the following five-number summary: Minimum = 10, Q1 = 40, Median = 50, Q3 = 55, Maximum = 60. Which of the following is the best description of the shape of the distribution?
Answer: Skewed to the left, because the distance from Q1 to the minimum is greater than the distance from Q3 to the maximum.
To assess skewness from a five-number summary, compare the distances between the components. The distance from Q1 to the minimum is 40 - 10 = 30. The distance from the maximum to Q3 is 60 - 55 = 5. Since the lower tail (Min to Q1) is much more spread out than the upper tail (Q3 to Max), the distribution is skewed to the left. Also, the distance from the median to Q1 (50 - 40 = 10) is greater than the distance from the median to Q3 (55 - 50 = 5), which also suggests left skewness.
- The boxplot summarizes the test scores for a class of students. Based on the boxplot shown:
Answer: The distribution of scores is skewed to the left, with no apparent outliers.
The boxplot shows a longer whisker on the left side, and the median is positioned closer to the third quartile (Q3) than the first quartile (Q1). The distance from the median to the minimum value is greater than the distance from the median to the maximum value. These features are all characteristic of a left-skewed distribution. There are no individual points plotted beyond the whiskers, which indicates there are no apparent outliers according to the 1.5 × IQR rule used to construct the plot.
- A real estate report for a large metropolitan area states that the mean home price is $780,000 and the median home price is $595,000. Which of the following statements is the most reasonable conclusion based on this information?
Answer: The distribution of home prices is likely skewed to the right.
In a skewed distribution, the mean is pulled in the direction of the long tail. When the mean is substantially greater than the median (mean > median), it indicates that there are high-value outliers or a long tail on the right side of the distribution. This pattern is characteristic of a right-skewed distribution, which is common for variables like income and housing prices.
- Two datasets, A and B, each with 100 data points, have the same mean of 50. The distribution of Dataset A is symmetric and unimodal (bell-shaped). The distribution of Dataset B is symmetric and bimodal, with peaks at 20 and 80. Which dataset is expected to have a larger standard deviation, and why?
Answer: Dataset B, because more of its data points are located far from the mean.
Standard deviation measures the typical distance of data points from the mean. In Dataset A (bell-shaped), most of the data is clustered near the mean of 50. In Dataset B (bimodal with peaks at 20 and 80), most of the data is clustered far from the mean of 50. Since more of the data in Dataset B lies at a greater distance from the mean, its standard deviation will be larger.
- A dataset consists of the official jersey numbers for the 23 players on a national soccer team roster. These numbers typically range from 1 to 23. Which of the following is the most likely shape of the distribution of these jersey numbers?
Answer: Approximately uniform
The jersey numbers will be the integers from 1 to 23, with each number appearing exactly once. A distribution where each value occurs with the same frequency (in this case, once) is a uniform distribution. There is no tendency for the numbers to cluster in the middle (as in a normal distribution) or at one end (as in a skewed distribution). Therefore, the shape will be approximately uniform.
