Indian Institute of Technology Madras
BS Degree Programme · Examination & Study Paper
BSMA1002 · Statistics for Data Science I
Qualifier · January 2024
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.
- Q1MCQ2 Marks·Choose one option

- AThere is an association between “Gender” and “Willingness” as row relative frequencies will be approximately same for all rows.
- BThere is no association between “Gender” and “Willingness” as row relative frequencies will be approximately same for all rows.
- CThere is an association between “Gender” and “Willingness” as row relative frequencies will be different for all rows.
- DThere is no association between “Gender” and “Willingness” as row relative frequencies will be different for all rows.
- Q2MSQ4 Marks·Choose all correct options

- AThere are 9 students who have scored greater than 20 marks.
- BThe data is bimodal.
- CMean is greater than the median.
- DInterquartile range is 28.
- Q3NAT4 Marks·Write the number
Your answer: - Q4NAT4 Marks·Write the number
Your answer: - Q5MSQ2 Marks·Choose all correct options
To ensure the accuracy of conclusions drawn through inferential statistics, which of the following statement(s) must be true?
- ASample should not be randomly selected.
- BSample should be randomly selected.
- CSample should be a good representative of the population.
- DSample should not be representative of the population.
- Q6MSQ2 Marks·Choose all correct options
Which of the following statement(s) is/are true?
- AThe gross annual income for each of 1000 randomly chosen households in New York City for the year 2000 is a time series data.
- BPassenger Name Record (PNR) number has an ordinal scale of measurement.
- CRevenue generated by India through tea exports to 10 different countries in year 2010 is a cross-sectional data.
- DShirt size is a categorical variable.
Read the passage below, then answer Q7–Q8
An analyst conducted a survey to understand the opinions of college students regarding the quality of education at their institution. He collected the data from 10 students as:
“Fair”, “Fair”, ”Excellent”, “Poor”, “Fair”, “Excellent”, “Poor”, “Poor”, “Excellent”, and “Poor”.
Based on the above data, answer the given subquestions.
Q7MCQ1 Mark·Choose one optionWhat is the mode of the given data?
- AExcellent
- BPoor
- CFair
- DMode is not defined for the given data
- Q8MCQ2 Marks·Choose one option
What is the median of the given data?
- APoor
- BFair
- CExcellent
- DMedian is not defined for given data
- Q9MSQ5 Marks·Choose all correct options

- AOf all the shirts with Polo style, there are 26.67% shirts of Medium size.
- BOf all the shirts with Small size, there are 25% shirts of Small Print style.
- CThere is no association between the size and style of shirts.
- DThere is an association between the size and style of shirts.
- Q10MCQ3 Marks·Choose one option

- A

- B

- C

- D

- A
- Q11MCQ3 Marks·Choose one option
Select the correct statements from the following:
- ATwo data sets with the identical frequency distributions will have identical relative frequency distributions.
- BA relative frequency is the number of observations belonging to a category.
- CTwo data sets with the identical relative-frequency distributions will always have identical frequency distributions.
- DIf all of the bars in a bar chart have the same length, then the categorical variable shown in the bar chart has no variation.
Read the passage below, then answer Q12–Q13
Q12NAT2 Marks·Write the numberFind the value of 10x.
Your answer:- Q13MSQ3 Marks·Choose all correct options
Choose the correct statement(s) from the following.
- AMode of the data is solar energy.
- BMedian of the data is wind energy.
- CThe combined consumption of the hydro, biomass and geothermal energy is more than the wind energy consumption.
- DThe consumption of solar energy is 120 kWh.
- Q14MSQ3 Marks·Choose all correct options

- APurchase and Region are categorical variables.
- BIncome Range has nominal scale of measurement.
- CAge has a ratio scale of measurement.
- DHousehold size is a discrete numerical variable.
- EPurchase has an ordinal scale of measurement.
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: B
- In the 100% stacked bars, about 22% of females and about 25% of males said No. The row relative frequencies are almost the same for both genders, so there is no association.
- Q2Answer: C, D
- Sum = 500, so mean = 500/20 = 25. Median = (10th + 11th)/2 = (23 + 25)/2 = 24, so mean > median. Q₁ = (5th + 6th)/2 = 10 and Q₃ = (15th + 16th)/2 = 38, so IQR = 28. Only 11 marks are above 20, and 10, 38, 50 each occur twice, so the data is not bimodal.
- Q3Answer: 39
- n = 10 + 4 + b + 4 + 1 + 2 = 21 + b. Sum = 4 + 2b + 12 + 4 + 10 = 30 + 2b. Mean: (30 + 2b)/(21 + b) = 1.6, so 0.4b = 3.6 and b = 9. Then n = 30, so b + n = 39.
- Q4Answer: 70
- x̄ = 350/5 = 70 and ȳ = 325/5 = 65. Deviations of X: −5, −5, −10, 10, 10; of Y: −5, −5, −10, 10, 10. Σ(products) = 25 + 25 + 100 + 100 + 100 = 350. Cov = 350/5 = 70.
- Q5Answer: B, C
- Inferential statistics uses a sample to draw conclusions about the population. For this to be accurate, the sample should be chosen at random and should represent the population well.
- Q6Answer: C, D
- Tea export revenue to 10 countries in one year is data at one point in time, so it is cross-sectional. Shirt size (S, M, L) is categorical. Incomes of households for one year are also cross-sectional, not time series, and a PNR number is only a label (nominal).
- Q7Answer: B
- Count each rating: Poor = 4, Fair = 3, Excellent = 3. Poor occurs most often, so the mode is Poor.
- Q8Answer: B
- The ratings are ordinal: Poor < Fair < Excellent. In order: P, P, P, P, F, F, F, E, E, E. The 5th and 6th values are both Fair, so the median is Fair.
- Q9Answer: A, D
- Polo total = 35 + 20 + 20 = 75, and Medium Polo = 20, so 20/75 = 26.67%. Small total = 100 and Small Print = 40, so it is 40%, not 25%. Row relative frequencies differ (Small 25/35/40%, Large 60/20/20%), so there is an association.
- Q10Answer: C
- Mean x̄ = 20/50 = 0.4. Population variance = 8/50 = 0.16, so σ = √0.16 = 0.4.
- Q11Answer: A
- Relative frequency = frequency ÷ total, so identical frequency tables give identical relative frequency tables. The reverse is false because the totals may differ (10 vs 100 items). A relative frequency is a proportion, not a count.
- Q12Answer: 357
- All slices add to 100%. x = 100 − (28.6 + 17.9 + 7.1 + 10.7) = 100 − 64.3 = 35.7. So 10x = 357.
- Q13Answer: A, C
- Solar has the largest share (35.7%), so it is the mode. Hydro + biomass + geothermal = 17.9 + 7.1 + 10.7 = 35.7% > 28.6% (wind). Energy source is nominal, so the median is not defined. Solar = 0.357 × 350 ≈ 125 kWh, not 120.
- Q14Answer: A, C, D
- Purchase (Yes/No) and Region are categorical. Age has a true zero, so it is ratio scale. Household size is a count, so it is discrete. Income range has an order, so it is ordinal, not nominal, and Purchase is nominal, not ordinal.