A list of ages recorded for a school trip includes the value 130 years. Write down why this value must be a data-entry error.
(Total for Question 1 is 1 mark)
2
A survey records a respondent's height as -160 cm. State why this value cannot be correct.
(Total for Question 2 is 1 mark)
3
Two rows in a database both list 'Age: 12, Name: Tom Lee, School: Elm Road'. Write down what type of data problem this is.
(Total for Question 3 is 1 mark)
4
A column of exam marks records one entry as 'ninety-two' while every other entry is a number such as 85. Write down the data-cleaning problem shown here.
(Total for Question 4 is 1 mark)
5
A list of 7 test scores out of 20 is: 12, 14, 13, 15, 41, 14, 13. Work out the range of the data as recorded, including the value 41.
(Total for Question 5 is 2 marks)
6
The value 41 in Question 5's list is later corrected to 14, since it was a data-entry error. Work out the range of the corrected data set: 12, 14, 13, 15, 14, 14, 13.
(Total for Question 6 is 2 marks)
7
A list of 8 delivery times, in minutes, is: 18, 21, 19, 22, 200, 20, 19, 21. Write down the value that is a data-entry error, and state what its correct value was most likely intended to be.
(Total for Question 7 is 2 marks)
8
A vet's spreadsheet has two entries for the same day: 'R. Ahmed, 14:00, Consultation' and 'Raj Ahmed, 14:00, Consultation'. Give two reasons why these are likely to be duplicate entries for the same appointment.
(Total for Question 8 is 2 marks)
9
A dataset of 6 room temperatures, in degrees C, is: 21, 22, 20, 23, 21, 2. Explain why 2 degrees C is unlikely to be a genuine room temperature reading, and suggest what digit was probably missed out when it was typed.
(Total for Question 9 is 2 marks)
10
A list of 7 shoe sizes is: 5, 6, 6, 7, 7, 7, 25. Work out the median shoe size, including the value 25.
(Total for Question 10 is 2 marks)
11
Using the same 7 shoe sizes from Question 10 (5, 6, 6, 7, 7, 7, 25), work out the mean shoe size.
(Total for Question 11 is 2 marks)
12
A spreadsheet records 5 delivery dates for the same week: 03/06/2027, 4/6/27, 05-06-2027, 2027-06-06, 07/06/2027. Give two reasons why these dates are difficult to compare directly.
(Total for Question 12 is 2 marks)
13
A gym's sign-in sheet lists 8 rows for one class, taken automatically from members scanning their cards. The Member IDs recorded are, in order: M501, M502, M501, M503, M504, M502, M505, M501. Work out how many unique members attended, and write down the Member ID that was recorded most often, stating how many times it appears.
(Total for Question 13 is 3 marks)
14
A researcher records the resting pulse rate, in beats per minute, of 8 volunteers: 66, 70, 68, 690, 71, 69, 67, 70. Identify the value that is almost certainly a data-entry error, explain why it cannot be a genuine pulse rate, and suggest the most likely correct value.
(Total for Question 14 is 3 marks)
15
The weekly earnings, in pounds, of 9 employees at a small cafe are: 310, 320, 305, 315, 300, 325, 318, 312, 980. One employee is the manager, who genuinely earns much more than the others (this is not an error). Work out the median weekly earnings and the mean weekly earnings for all 9 employees, and state which average better represents the typical earnings, giving a reason.
(Total for Question 15 is 3 marks)
16
A teacher records the time, in minutes, that 10 students take to complete a test: 32, 35, 33, 34, 36, 33, 35, 34, 32, 95. Give a reason why 95 minutes might be a genuine time rather than a data-entry error, state one appropriate action the teacher could take with this value if it is confirmed genuine, and explain why removing a confirmed genuine value would not be appropriate.
(Total for Question 16 is 3 marks)
17
A clinic's records show the weight, in kg, of 10 newborn babies. One value is missing: 3.2, 3.4, 3.1, 3.5, 3.3, 3.6, 3.0, 3.4, (missing), 3.2. Give one appropriate way the clinic could estimate the missing weight, then use the mean of the other 9 weights to work out this estimate, and state one disadvantage of simply deleting this baby's record instead.
(Total for Question 17 is 4 marks)
18
The number of emails received by 11 employees in one day, given in ascending order, is: 8, 9, 9, 10, 11, 11, 11, 12, 13, 14, 40. Find the lower quartile (Q1) and upper quartile (Q3) of the data, work out the interquartile range (IQR), and then use the rule that a value is an outlier if it is higher than Q3 + 1.5 x IQR to decide whether 40 is an outlier.
(Total for Question 18 is 4 marks)
Mark scheme · F01D Data Cleaning and Outliers (Foundation): Fluency and Exam Drill
Question 1
B1 130 years is an impossible age for a person, so it must be an error oe
Answer: 130 years is an impossible age for a person (nobody lives that long), so it must be a data-entry error.
Question 2
B1 a height cannot be negative oe
Answer: A height cannot be negative, so -160 cm is impossible and must be an error.
Question 3
B1 duplicate entry / the same record entered twice oe
Answer: A duplicate entry: the same record has been entered twice.
Question 4
B1 the value is recorded as text/words instead of a number, so it is in an inconsistent format oe
Answer: The value is written as words instead of a number, so it is in an inconsistent format and will not be read as numeric data.
Question 5
M1 41 - 12 oe
A1 29 cao
Answer: 29
Question 6
M1 15 - 12 oe
A1 3 cao
Answer: 3
Question 7
B1 200 identified as the error cao
B1 20 stated as the likely intended value oe
Answer: 200 is the error. The intended value was most likely 20 (an extra zero was typed by mistake).
Question 8
B1 both entries share the same time, 14:00 oe
B1 'R. Ahmed' is a shortened form of 'Raj Ahmed' oe
Answer: Both entries share the same time (14:00), and 'R. Ahmed' is simply a shortened version of the name 'Raj Ahmed'.
Question 9
B1 2 degrees C is far too cold for an indoor room, unlike the rest of the data (20 to 23) oe
B1 a leading '2' was probably missed off, so the true reading was likely 22 oe
Answer: 2 degrees C is far too cold to be a genuine indoor room temperature; a leading digit was probably dropped, so the true value was likely 22 degrees C.
Question 10
M1 identifies the data is already in order and finds the middle (4th) value oe
A1 7 cao
Answer: 7
Question 11
M1 (5+6+6+7+7+7+25) / 7 oe
A1 9 cao
Answer: 9
Question 12
B1 different orderings are used, e.g. day/month/year and year-month-day oe
B1 inconsistent use of leading zeros and full/abbreviated years oe
Answer: The dates use different orderings (some day/month/year, one year-month-day) and inconsistent leading zeros and year lengths, so they cannot be compared or sorted reliably as they are.
Question 13
M1 groups the 8 rows by Member ID oe
A1 5 unique members cao
B1 M501, recorded 3 times, cao
Answer: 5 unique members; M501 was recorded most often, 3 times.
Question 14
B1 690 identified as the error cao
B1 explains 690 bpm is far higher than any realistic resting pulse rate oe
B1 69 stated as the most likely correct value cao
Answer: 690 is the error; a resting pulse of 690 bpm is impossible for a human, so an extra digit was probably typed. The correct value was most likely 69 bpm.
Question 15
B1 median = 315 cao
B1 mean = 387.22 (awrt) cao
B1 median is more representative, since the manager's high (genuine) salary pulls the mean up oe
Answer: Median = 315 pounds, mean = 387.22 pounds (awrt); the median better represents typical earnings because the manager's high salary pulls the mean up.
Question 16
B1 reason 95 minutes could be genuine, e.g. the student may have needed extra time or has additional learning needs oe
B1 appropriate action, e.g. keep the value and note it separately rather than deleting it oe
B1 removing a genuine value would lose real information and make the data set less representative oe
Answer: 95 minutes could genuinely reflect a student needing extra time; the teacher should keep the value (noting it separately) rather than delete it, since removing a confirmed genuine value would lose real information and make the data less representative.
Question 17
B1 sensible method, e.g. estimate using the mean of the other 9 weights, or check the hospital notes oe
M1 (3.2+3.4+3.1+3.5+3.3+3.6+3.0+3.4+3.2) / 9 oe
A1 3.3 kg cao
B1 disadvantage of deleting, e.g. loses information about that baby and reduces the sample size oe
Answer: Estimate the missing weight using the mean of the other 9 weights, which is 3.3 kg; deleting the record instead would lose information about that baby and reduce the sample size.
Question 18
B1 Q1 = 9 cao
B1 Q3 = 13 cao
B1 IQR = 4 (ft from Q3 - Q1)
B1 boundary = 19 (ft), and 40 correctly identified as an outlier since 40 > 19
Answer: Q1 = 9, Q3 = 13, IQR = 4, upper boundary = 19; since 40 is greater than 19, it is an outlier.