68 practice questions on Statistics , sorted Easy → Hard. Try each one first, then open its answer page for the worked explanation. Want the full theory first? Read the Statistics notes .
Normal Distribution: the 68-95-99.7 Rule 68% within ±1σ 95% within ±2σ mean=median=mode (centre) In a perfectly normal (bell-shaped) distribution, mean, median, and mode all coincide at the centre; about 68% of data falls within 1 standard deviation of the mean, 95% within 2, and 99.7% within 3 - the empirical rule used to judge how typical or extreme a value is.
Easy - 20 questions Q9.
Data presented as a bar graph shows:
A Continuous data measured across a rangeB Frequencies of different categoriesC Change occurring gradually over timeD Probability values assigned to outcomesShow answer & explanation →
Q10.
What does a histogram show?
A Categorical data plotted as separate, non-adjacent barsB Frequency distribution of continuous data in intervalsC A circular sector representation equivalent to a pie chartD Only the averages of each class intervalShow answer & explanation →
Q12.
Standard deviation measures:
A The central tendency of the data, like the mean doesB Spread (variability) of data around the meanC The middle value of the data when arranged in orderD The single largest value present in the data setShow answer & explanation →
Medium - 20 questions Q21.
Coefficient of variation is used to:
A Find the mean value of a single data setB Compare variability between datasets with different meansC Measure the correlation between two distinct variablesD Find the quartile boundaries of a frequency distributionShow answer & explanation →
Q22.
The median of grouped data uses the formula:
A l + (n/2 - F)/f × hB l + (n - F)/f × hC l + (f₁ - f₀)/(2f₁ - f₀ - f₂) × hD sigma(fx)/sigma(f)Show answer & explanation →
Q23.
In grouped data, the mode formula is:
A l + [(f₁-f₀)/(2f₁-f₀-f₂)] × hB l + (n/2-F)/f × hC Mean - Mode = 3(Mean-Median)D sigma(fx)/nShow answer & explanation →
Q24.
Pearson's correlation coefficient r = 1 means:
A No correlationB Perfect negative correlationC Perfect positive correlationD Some positive correlationShow answer & explanation →
Q25.
IQR (Interquartile Range) = Q<sub>3</sub> - Q<sub>1</sub> measures:
A Total range from minimum to maximumB Spread of middle 50% of dataC Average spread across the entire datasetD Variance, the squared average deviationShow answer & explanation →
Q26.
Standard deviation is preferred over mean deviation because:
A It is easier to calculate since it largely avoids using absolute valuesB It gives more weight to extreme values and has algebraic tractabilityC It tends to be smaller in value than the mean deviation in practiceD It equals the variance directly, making the two values interchangeableShow answer & explanation →
Q27.
If all values in a dataset are multiplied by 2, the standard deviation:
A Stays the sameB DoublesC Is halvedD Increases by 2Show answer & explanation →
Q31.
If adding a constant k to each observation, mean:
A Remains the sameB Increases by kC Multiplies by kD Becomes 0Show answer & explanation →
Q32.
The empirical relation between mean, median, and mode is:
A Mode = Mean + 2 MedianB Mean - Mode = 3(Mean - Median)C Mode = 2 Mean - 3 MedianD Median = Mean - ModeShow answer & explanation →
Q33.
For a negatively skewed distribution:
A Mean = Median = ModeB Mean > Median > ModeC Mean < Median < ModeD Mode < Median < MeanShow answer & explanation →
Q34.
Step-deviation method for mean uses:
A Direct formula applying raw class marks without taking any shortcutB Simplification by subtracting assumed mean and dividing by class width (h)C Decomposition of the frequency expression into separate partial fractionsD Taking the logarithm of each class mark before averaging the resultShow answer & explanation →
Q35.
Spearman's rank correlation is used when:
A Data is normally distributed with a known population variance valueB Data is on ordinal scale or ranks, or non-linear relationship suspectedC Data has roughly equal variance across the observed groupsD Used mainly for large samples exceeding thirty observationsShow answer & explanation →
Q36.
The box-plot whiskers extend to:
A Mean plus or minus two standard deviationsB Minimum and maximum values (or 1.5×IQR from Q<sub>1</sub>/Q<sub>3</sub>)C Exactly the first and third quartiles, Q<sub>1</sub> and Q<sub>3</sub>D Mean plus or minus one standard deviationShow answer & explanation →
Hard - 28 questions Q41.
For grouped data, Pearson's first coefficient of skewness is:
A (Mean - Mode)/SDB (Mean - Median)/SDC 3(Mean - Median)/SDD (Median - Mode)/SDShow answer & explanation →
Q42.
The regression line y on x passes through:
A The origin point specifically, regardless of the dataB (x̄, ȳ) - the meansC All of the individual data points at onceD Just the first and last data point in the setShow answer & explanation →
Q43.
If byx × bxy = r², where byx and bxy are regression coefficients, then r =
A byx × bxyB sqrt(byx × bxy)C (byx + bxy)/2D byx/bxyShow answer & explanation →
Q45.
Partial correlation measures:
A Correlation removing effect of a third variableB The simple two-variable correlation with no adjustment appliedC The standardized coefficient from a multiple regression modelD The cross-correlation between two time-shifted signalsShow answer & explanation →
Q46.
The variance of combined data of two groups (n₁, x̄₁, σ₁²) and (n₂, x̄₂, σ₂²) is:
A (n₁σ₁²+n₂σ₂²)/(n₁+n₂), which overlooks the shift in means between the two groupsB A more complex formula involving d₁² and d₂² (distances from combined mean)C (σ₁²+σ₂²)/2, a simple unweighted average that ignores group sizesD σ₁ × σ₂, the simple product of the two individual standard deviationsShow answer & explanation →
Q47.
Moment generating function M(t) of a random variable X is defined as:
A E[X]B E[X²]C E[e<sup>tX</sup>]D E[tX]Show answer & explanation →
Q48.
Stratified random sampling ensures:
A Every individual in the population has an exactly equal chanceB Proportional representation from each subgroup (stratum)C Only random selection, with strata playing no role at allD Equal sample size drawn from every group, regardless of its sizeShow answer & explanation →
Q50.
Laspeyres price index uses base year quantities. It tends to:
A Underestimate inflationB Overestimate inflationC Give exact inflationD Equal Paasche index alwaysShow answer & explanation →
Q51.
In regression analysis, R² (coefficient of determination) measures:
A The correlation coefficient itself, without squaring itB Proportion of variance in Y explained by XC The slope of the fitted regression lineD The raw sum of squared residuals from the regressionShow answer & explanation →
Q59.
The standard deviation of a data set is the ___ of its variance:
A the square rootB the squareC the reciprocalD double the valueShow answer & explanation →
Q62.
If every observation in a data set is increased by 5, the variance:
A increases by 5B increases by 25C remains unchangedD increases by 10Show answer & explanation →
Q63.
The mean of 100 observations is 50. Later one observation 100 was found to be misread as 10. The corrected mean is:
Show answer & explanation →