Indian Institute of Technology Madras
BS Degree Programme · Examination & Study Paper
BSMA1002 · Statistics for Data Science I
All 2023 papers
Examination Instructions & Guidelines
- Solve on notebook/paper first before checking solutions.
- MCQ: Exactly 1 option is correct. Negative marking applies.
- MSQ: One or more options correct. Select all to score.
- NAT: Numerical response. No negative marking.
- Scientific calculator allowed for all calculations.
- Official answer key is sealed at the end of this sheet.
Qualifier · September 2023Qualifier
- Q1NAT3 Marks·Write the number
Your answer: Read the passage below, then answer Q2–Q3
The Head of the transportation facility at the IIT Madras campus wants to conduct a survey to rate the bus services offered to all the students. He randomly selects 100 final year students from the campus. Based on the given information, answer the subquestions
Q2MSQ3 Marks·Choose all correct optionsIdentify the sample and the population.
- AThe population is all the students at IIT Madras campus.
- BThe sample is all the final year students at IIT Madras campus.
- CThe sample is 100 randomly selected final year students at IIT Madras campus.
- DThe population is all the final year students at IIT Madras campus.
- Q3MCQ2 Marks·Choose one option
Identify the correct statement from the following.
- ASelected sample is a good representation of the population.
- BSelected sample is not a good representation of the population.
Read the passage below, then answer Q4–Q5
Q4NAT2 Marks·Write the numberIf a total of 500 fruits are sold on a given day, then find the number of mangoes sold.
Your answer:- Q5MSQ3 Marks·Choose all correct options
Choose the correct statement(s) from the following.
- AThe slices of pie chart adds up to 100%.
- BMedian of the data is Grapes.
- CMode of the data is Guava.
- DThe pie chart is misleading because it does not obey the area principle.
Read the passage below, then answer Q6–Q7
Q6NAT5 Marks·Write the numberCalculate the absolute value of Point Bi-serial correlation coefficient between the result and number of hours spent. Enter the answer correct to three decimal places.
Your answer:- Q7MCQ2 Marks·Choose one option
Are the exam results of the selected students strongly influenced by the number of hours spent studying?
- AYes, because the absolute value of Point Bi-serial coefficient is close to 1.
- BNo, because the absolute value of Point Bi-serial coefficient is close to 1.
- CYes, because the absolute value of Point Bi-serial coefficient is close to 0.
- DNo, because the absolute value of Point Bi-serial coefficient is close to 0.
Read the passage below, then answer Q8–Q9
Q8NAT3 Marks·Write the numberWhat is the average annual package of the students if each observation is doubled after subtracting 5 from it?
Your answer:- Q9MCQ4 Marks·Choose one option
What is the population variance of the given data if 5 is subtracted from each observations?
- A78.6
- B83.6
- C88.6
- D93.6
- Q10NAT4 Marks·Write the number
Your answer: - Q11MSQ2 Marks·Choose all correct options
Which of the following statement(s) is(are) true?
- AEye colour of a person has ordinal scale of measurement.
- BRecording the price of bitcoin every hour for 3 days is a time-series data.
- CNumber of students present in a class is a discrete variable.
- DAmount of milk (in litres) is a categorical variable.
- Q12MSQ2 Marks·Choose all correct options
Sunidhi purchased 5 t-shirts of sizes M, L, S, XL, S for her cousins. Later she purchased one more t-shirt of size M, then choose the correct option(s) from the following:
- AMedian of the new data is same as of old data.
- BNew data will be bimodal.
- CMedian of the new data will change.
- DNew data will be unimodal.
- Q13MSQ3 Marks·Choose all correct options

- A

- B

- C

- D

- A
- Q14NAT2 Marks·Write the number
Find the IQR (Interquartile range) of the data 14, 20, 40, 23 and 18.
Your answer:
Qualifier · May 2023Qualifier
- Q15MCQ2 Marks·Choose one option
A researcher wants to investigate the relationship between studying hours and exam scores. He collects the data on studying hours and exam scores from a sample of 100 students and perform a correlation analysis. Which branch of statistics does this analysis belong to?
- ADescriptive Statistics.
- BInferential Statistics.
- CBoth Descriptive and Inferential Statistics.
- DNeither Descriptive nor Inferential Statistics.
- Q16MSQ2 Marks·Choose all correct options

- AIt is a time series Data.
- BIt is a cross sectional Data.
- CIt is a structured Data.
- DIt is an unstructured Data.
- Q17MSQ2 Marks·Choose all correct options
Which of the following is/are correct?
- AMean and Mode can only be defined for a numerical data.
- BMode can be defined for a categorical data.
- CThere can be more than one mode for a data.
- DMedian can be measured for a categorical variable only if the variable is nominal.
- EThe sum of relative frequencies of all the observations in a given data set is always equal to 1.
- Q18NAT4 Marks·Write the number
Your answer: - Q19NAT4 Marks·Write the number
The mean and sample variance of the data set consisting of 10 observations is 22 and 81 respectively. Later it is noted that one observation 20 is wrongly noted as 15. what is the sample standard deviation of the original data set?(Enter the answer correct to two decimal accuracy)
Your answer: Read the passage below, then answer Q20–Q22
Table 2 represents the movies released in 2022 on the OTT platform.

Based on the information, answer the given subquestions.
Q20NAT2 Marks·Write the numberFind the total number of released movies in 2022 on OTT Platform.
Your answer:- Q21NAT2 Marks·Write the number
Find the number of Thriller and Comedy movies released in 2022 on OTT Platform.
Your answer: - Q22NAT3 Marks·Write the number
Calculate a + b. (Enter the answer correct to three decimal accuracy)
Your answer: Read the passage below, then answer Q23–Q24
Based on the given data in the Table 3 , answer the subquestions.
Q23MSQ2 Marks·Choose all correct optionsWhich of the following statements is/are true?
- AXYZ Ltd. is a variable.
- BNumber of employees have a ratio scale of measurement.
- CGrowth Rate (%) is a variable.
- DCategory have an interval scale of measurement.
- Q24MCQ2 Marks·Choose one option
What is the scale of measurement of “Category”?
- AOrdinal Scale.
- BRatio Scale.
- CNominal Scale.
- DInterval Scale.
Read the passage below, then answer Q25–Q26
Table 5 represents the data set about Advertising Expenditure and the Sales Revenue of a company. An analyst wants to investigate the relationship between the Advertising Expenditure (in Lakhs of rupees) and the monthly Sales Revenue (in Lakhs of rupees) of a company over a period of 6 months.

Based on the information, answer the given subquestions.
Q25MSQ2 Marks·Choose all correct optionsWhich measure(s) will help you to determine the existence of the relation between Sales Revenue and the Advertising Expenses?
- ACovariance
- BMean of both Sales and Advertising
- CCorrelation Coefficient
- DNone
- Q26NAT4 Marks·Write the number
What is the sample covariance between the Sales Revenue and Advertising Expenses?(Enter the answer correct to one decimal accuracy)
Your answer: Read the passage below, then answer Q27–Q29
A clothe retailer wants to analyze the prices of different brands of t-shirts in his store. He randomly collected 10 t-shirts and prices (in rupees) of the t-shirts are 150, 550, 700, 240, 300, 750, 200, 180, 320, 420 respectively.
Based on the information, answer the given subquestions.
Q27MSQ3 Marks·Choose all correct optionsPrice below which 25% of the t-shirts fall.
- AIt is a measure of Q₁.
- B200 Rs.
- CIt is a measure of Q₃.
- D550 Rs.
- Q28MSQ3 Marks·Choose all correct options
Price of the t-shirt that separates price of the highest 25% of the data from the lower 75%.
- AIt is a measure of Q₂
- B550 Rupees
- CIt is a measure of Q₃
- D310 Rupees
- Q29MCQ3 Marks·Choose one option
How many outliers are in the given data set?
- A0
- B1
- C2
- DCan’t determined.
Qualifier · January 2023Qualifier
Read the passage below, then answer Q30–Q31
Figure Q.1 represents the placement report of an engineering college for the academic year 2021-22.

Based on the information, answer the given subquestions.
Q30MCQ2 Marks·Choose one optionWhich of the following category does not represent the mode?
- AAnalytics
- BSoftware
- CCore
- DInsufficient information
- Q31NAT3 Marks·Write the number
If 1200 students are placed during the placement season of 2021-22, then find the number of students placed in core companies?
Your answer: - Q32MSQ3 Marks·Choose all correct options
Which of the following statements is/are true?
- AA sample is the subset of a population.
- BNumerical variables can have all the properties of ordinal and nominal scales of measurement.
- CCategorical variable can have numerical properties.
- DCorrelation coefficient measures a linear association between two numerical variables.
- Q33MSQ3 Marks·Choose all correct options
Which of the following represent(s) cross sectional data?
- ACensus of tigers estimated in four different countries in 2021.
- BYearly irrigation report released by a government over the period of ten years.
- CRevenue generated by a company in four different states in 2022.
- DMonthly expenditure of a family in 2020.
- Q34MSQ3 Marks·Choose all correct options
Choose the correct statement/s ?
- AA nominal scale has the property of labelling the categories and it does not involve the ranking of data.
- BAn ordinal scale has all the properties of nominal scale and it involves the ranking of data.
- CAn Interval scale has all the properties of ordinal scale and it satisfies the absolute zero property.
- DA ratio scale has all the properties of interval scale and it does not satisfy the absolute zero property.
- Q35MCQ4 Marks·Choose one option
Suppose the number of cars sold by a car dealer in five months are 12, 8, 9, 20, 40. Which of the following is/are outliers?
- A-8
- B20
- C40
- D8
- Q36MSQ4 Marks·Choose all correct options
An analyst did a survey to know whether literacy rate affects voting rate among adults in villages. He collected the data from a village and the results are tabulated in Table Q.1.
Choose the correct option/s?- AOf all the literate adults, 25% do not participate in voting.
- BOf all the illiterate adults, 72.73% participate in voting.
- CIf all row relative frequencies are similar within each column then, it implies that all column relative frequencies will also be similar within each row.
- DIf all column relative frequencies are similar within each row then, it does not imply that all row relative frequencies will also be similar within each column.
- EThere is no association between literacy and voting participation.
- FThere is an association between literacy and voting participation.
Read the passage below, then answer Q37–Q38
An analyst want to analyse the salary of employees in different organizations in a city. To analyse this, he has selected an organization and the data of salaries is tabulated as shown in Table Q.2.

Based on the given information, answer the subquestions.
Q37MCQ2 Marks·Choose one optionBased on the data collected from an organisation, analyst made a statement that the average salary of an employee is 50,000 rupees in different organizations in the city. The given statement of analyst is based on which kind of statistical analysis ?
- ADescriptive statistics
- BInferential statistics
- Q38NAT4 Marks·Write the number
What is the sample standard deviation of salary (in thousand rupees)? (Enter the answer correct to 2 decimal accuracy)
Your answer: Read the passage below, then answer Q39–Q40
The heights (in cm) of ten athletes are represented by a stem and leaf plot in Figure Q.2.

Based on the above information, answer the given subquestions
Q39NAT3 Marks·Write the numberWhat is the median height(in cm) of an athlete?
Your answer:- Q40NAT3 Marks·Write the number
What is the mode of the heights (in cm) of athletes?
Your answer: - Q41NAT3 Marks·Write the number
The mean and sample variance of the marks (out of 50) scored by the students in the Statistics course are 40 and 25 respectively. If 5 is subtracted from the marks of each student and then multiplied by 2. Find the mean of the modified marks.
Your answer: - Q42MCQ3 Marks·Choose one option
Let X and Y represent the total marks of Mayur and Deepak in 5 subjects respectively. If the marks scored by Deepak is twice the marks scored by Mayur in each subject, then which of the following statement is correct?
- AThere is no correlation between X and Y .
- BThere is a weak positive correlation between X and Y .
- CThere is a strong positive correlation between X and Y .
- DInsufficient information.
Official KeyMaster Answer Key & Explanations
Official answer keys from IIT Madras BS past papers. Click any question to review.
Reveal KeyCollapse Key
Master Answer Key & Explanations
Official answer keys from IIT Madras BS past papers. Click any question to review.
Detailed Answer Notes & Working
- Q1Answer: 10
- Population variance = (3√2)² = 18 and sample variance = (2√5)² = 20. Since n × 18 = (n − 1) × 20, we get 20 = 2n, so n = 10.
- Q2Answer: A, C
- The survey is about bus services for all students, so the population is all students at IIT Madras. The sample is the 100 final year students who were randomly selected.
- Q3Answer: B
- The population is all students, but the sample has only final year students. Other years are left out, so the sample is not a good representation of the population.
- Q4Answer: 125
- Mango has 25% of the pie. Mangoes sold = 25% of 500 = 0.25 × 500 = 125.
- Q5Answer: A, C
- The slices 10 + 15 + 20 + 25 + 30 = 100%. Guava has the largest share (30%), so it is the mode. Fruit type is nominal, so the median is not defined, and the slice areas match the percentages, so the chart is not misleading.
- Q6Answer: 0.04 to 0.05
- Pass hours: 20, 25, 13 (mean 19.33). Fail hours: 19, 18, 20 (mean 19). Population SD of all 6 hours = √(74.83/6) ≈ 3.53. r = (19.33 − 19)/3.53 × √(3/6 × 3/6) ≈ 0.047. Official answer key accepts any value from 0.04 to 0.05.
- Q7Answer: D
- The point bi-serial coefficient is about 0.047, which is very close to 0. So the result is not strongly influenced by the number of hours spent.
- Q8Answer: 40
- Data: 10, 12, 16, 25, 25, 25, 30, 32, 34, 41. Sum = 250, so mean = 25. New value = 2(x − 5), so new mean = 2 × (25 − 5) = 40.
- Q9Answer: C
- Subtracting 5 from every value does not change the variance. Mean = 25. Squared deviations: 225, 169, 81, 0, 0, 0, 25, 49, 81, 256, sum = 886. Population variance = 886/10 = 88.6.
- Q10Answer: 0.29 to 0.35
- Means: Chemistry 30, Biology 25. Deviations: (−10, −5, 0, 5, 10) and (0, −5, 5, −5, 5). Σ(products) = 50, Σx² = 250, Σy² = 100. r = 50/√(250 × 100) = 50/158.1 ≈ 0.32. Official answer key accepts any value from 0.29 to 0.35.
- Q11Answer: B, C
- Bitcoin price recorded every hour is data over time, so it is time series. Number of students is a count, so it is discrete. Eye colour has no order, so it is nominal, and amount of milk in litres is numerical (continuous), not categorical.
- Q12Answer: A, B
- Old data in order: S, S, M, L, XL, so the median is M. New data: S, S, M, M, L, XL. The 3rd and 4th values are both M, so the median is still M. S and M now appear twice each, so the data is bimodal.
- Q13Answer: B, D
- m₂ − m₁ = [(x₂ + … + xₙ₊₁) − (x₁ + … + xₙ)]/n = (xₙ₊₁ − x₁)/n. In the same way, m₃ − m₂ = (xₙ₊₂ − x₂)/n. So options B and D are true.
- Q14Answer: 5
- Sorted data: 14, 18, 20, 23, 40 (n = 5). Q₁: 0.25 × 5 = 1.25, round up to the 2nd value = 18. Q₃: 0.75 × 5 = 3.75, round up to the 4th value = 23. IQR = 23 − 18 = 5.
- Q15Answer: A
- The correlation is calculated only to describe the relation within the collected data of 100 students. No conclusion is drawn about a larger population, so this is descriptive statistics.
- Q16Answer: A, C
- Oil production is recorded quarter by quarter over time, so it is time series data. It is arranged in a table with fixed rows and columns, so it is structured data.
- Q17Answer: B, C, E
- Mode can be found for categorical data (most frequent category), and a data set can have more than one mode. Relative frequencies always add up to 1. Median needs an order, so it works for ordinal data, not nominal.
- Q18Answer: 5
- n = (x + 5) + 15 + 20 + x = 2x + 40. Sum = 3(x + 5) + 60 + 20(x + 1) + x(x + 3) = x² + 26x + 95. Set sum = 5n: x² + 26x + 95 = 10x + 200, so x² + 16x − 105 = 0, (x + 21)(x − 5) = 0. So x = 5.
- Q19Answer: 8.68 to 8.71
- Σx = 220, Σ(x − x̄)² = 9 × 81 = 729, so Σx² = 729 + 10 × 22² = 5569. Correct values: Σx = 225, Σx² = 5569 − 15² + 20² = 5744, mean = 22.5. s² = (5744 − 10 × 22.5²)/9 = 681.5/9 ≈ 75.72, so s ≈ 8.70. Official answer key accepts any value from 8.68 to 8.71.
- Q20Answer: 200
- Action has 54 movies with relative frequency 0.27. Total = 54/0.27 = 200 movies.
- Q21Answer: 37
- Total = 54/0.27 = 200. Comedy d = 0.085 × 200 = 17 and Thriller c = 0.1 × 200 = 20. So Thriller + Comedy = 17 + 20 = 37.
- Q22Answer: 0.542 to 0.548
- Total = 54/0.27 = 200. a = 68/200 = 0.34 and b = 41/200 = 0.205. So a + b = 0.545. Official answer key accepts any value from 0.542 to 0.548.
- Q23Answer: B, C
- Number of employees has a true zero and ratios make sense, so it is ratio scale. Growth rate is a column that changes from stock to stock, so it is a variable. XYZ Ltd. is one observation, not a variable, and Category is nominal.
- Q24Answer: C
- Category (Technology, Healthcare, Energy, …) only names groups. There is no order between them, so it has a nominal scale.
- Q25Answer: A, C
- Covariance shows the direction of the relation between two numerical variables, and the correlation coefficient shows its direction and strength. The means alone say nothing about how the two variables move together.
- Q26Answer: 18.6 to 19.2
- Mean of Advertising = 117/6 = 19.5, mean of Sales = 495/6 = 82.5. Σ(products of deviations) = −1.25 + 33.75 + 41.25 + 0.75 + 13.75 + 6.25 = 94.5. Sample covariance = 94.5/5 = 18.9. Official answer key accepts any value from 18.6 to 19.2.
- Q27Answer: A, B
- The value below which 25% of data falls is Q₁. Sorted: 150, 180, 200, 240, 300, 320, 420, 550, 700, 750. 0.25 × 10 = 2.5, round up to the 3rd value, so Q₁ = 200 Rs.
- Q28Answer: B, C
- The value that separates the top 25% from the lower 75% is Q₃. Sorted data has 10 values. 0.75 × 10 = 7.5, round up to the 8th value, so Q₃ = 550 Rs.
- Q29Answer: A
- Q₁ = 200 and Q₃ = 550, so IQR = 350 and 1.5 × IQR = 525. Fences: 200 − 525 = −325 and 550 + 525 = 1075. All prices (150 to 750) are inside, so there are 0 outliers.
- Q30Answer: C
- Analytics and Software both have the highest share (30%), so both are modes. Core has 25%, so it does not represent the mode.
- Q31Answer: 300
- Core companies have 25% of the placements. Number placed in core = 25% of 1200 = 0.25 × 1200 = 300.
- Q32Answer: A, B, D
- A sample is a part (subset) of the population. Numerical variables can be labelled and ordered, so they have nominal and ordinal properties too. Correlation measures linear association between two numerical variables. Categorical variables do not have numerical properties.
- Q33Answer: A, C
- Cross-sectional data is collected at one point in time across many subjects. Tiger census in 4 countries in 2021 and revenue in 4 states in 2022 fit this. A ten-year report and monthly expenses are recorded over time, so they are time series.
- Q34Answer: A, B
- A nominal scale only labels categories with no ranking. An ordinal scale has nominal properties plus ranking. An interval scale has no absolute zero, and a ratio scale does have an absolute zero, so C and D are false.
- Q35Answer: C
- Sorted: 8, 9, 12, 20, 40. Q₁ = 2nd value = 9 and Q₃ = 4th value = 20, so IQR = 11 and 1.5 × IQR = 16.5. Upper fence = 20 + 16.5 = 36.5. Only 40 is above it, so 40 is the outlier.
- Q36Answer: A, B, C, E
- Literate: 10 of 40 do not vote = 25%. Illiterate: 40 of 55 vote = 72.73%. Literate adults vote 75%, which is close to 72.73%, so there is no association. If row relative frequencies are similar, column relative frequencies are also similar, so C is true and D is false.
- Q37Answer: B
- The analyst used data from only one organisation and made a claim about all organisations in the city. Drawing a conclusion about a population from a sample is inferential statistics.
- Q38Answer: 12.22 to 12.28
- Mean = (80 + 60 + 50 + 60 + 50)/5 = 60. Squared deviations: 400, 0, 100, 0, 100, sum = 600. Sample variance = 600/4 = 150, so s = √150 ≈ 12.25. Official answer key accepts any value from 12.22 to 12.28.
- Q39Answer: 175
- Heights in order: 162, 162, 168, 169, 175, 175, 175, 177, 180, 184. There are 10 values, so median = (5th + 6th)/2 = (175 + 175)/2 = 175 cm.
- Q40Answer: 175
- In the plot, 175 appears three times (17 | 5 5 5). No other height appears more than twice, so the mode is 175 cm.
- Q41Answer: 70
- Modified mark = 2(x − 5). The mean changes in the same way: new mean = 2 × (40 − 5) = 2 × 35 = 70.
- Q42Answer: C
- Y = 2X for every subject, which is a perfect positive linear relation. So the correlation is r = 1, a strong positive correlation.