- Home
- /
- Statistics
- /
- Mistakes
- /
- Center, Spread & Distributions
Common Mistakes in Center, Spread & Distributions
76 common pitfalls across 19 concepts, including 19 recurring stuck points
These are the pitfalls students hit across all 19 center, spread & distributions concepts, from Mean as Fair Share, Median and Mode onward. Each entry names the concept it comes from, so you can tell a slip from a genuine gap and go straight to the explanation.
Students often know a procedure related to mean as fair share but skip the recognition step: Would pooling all the amounts and sharing them out evenly give the one number you need? That leads to a calculation or graph that looks reasonable but answers a different question.
Thinking someone actually scored the mean
Using mean when extreme values distort it
Dividing by the number of categories instead of the number of values
Students often know a procedure related to median but skip the recognition step: Do you need the value in the middle position once the data are sorted, so that half lie on each side? That leads to a calculation or graph that looks reasonable but answers a different question.
Forgetting to order data first
Confusing with mean
Not averaging two middle values for even-sized data
Students often know a procedure related to mode but skip the recognition step: Do you need the most common value or category, rather than an even share or a middle value? That leads to a calculation or graph that looks reasonable but answers a different question.
Thinking there must always be one mode
Confusing mode with mean
Not recognizing bimodal data
Students often know a procedure related to range but skip the recognition step: Do you need the gap between the very largest and very smallest values, and nothing about the values in between? That leads to a calculation or graph that looks reasonable but answers a different question.
Forgetting to subtract
Confusing with number of values
Ignoring that outliers inflate range
Students often know a procedure related to mean vs median but skip the recognition step: Must you choose between the mean and the median, because a few extreme values might pull one of them away from typical? That leads to a calculation or graph that looks reasonable but answers a different question.
Always using mean without checking for outliers
Thinking higher average is always better
Forgetting that for skewed data the mean and median can be very different
Students often know a procedure related to spread vs center but skip the recognition step: Do the groups share a typical value yet differ in how widely their values scatter, so a center alone misleads? That leads to a calculation or graph that looks reasonable but answers a different question.
Reporting only center
Assuming same center means similar data
Ignoring spread in comparisons
Students often know a procedure related to data variability but skip the recognition step: Is the question whether the values stay close together or scatter widely, without asking you to calculate a spread number? That leads to a calculation or graph that looks reasonable but answers a different question.
Thinking same average means same data
Ignoring spread when comparing groups
Reporting only the mean without any measure of variability
Students often know a procedure related to quartiles but skip the recognition step: Do you need the three cut points that split sorted data into four groups with equal numbers of values? That leads to a calculation or graph that looks reasonable but answers a different question.
Including median when calculating /
Reading as the 1st percentile instead of the 25th
Not ordering data first
Students often know a procedure related to interquartile range (iqr) but skip the recognition step: Do you need how spread out the middle half of the data is, ignoring the lowest and highest quarters? That leads to a calculation or graph that looks reasonable but answers a different question.
Confusing with range
Calculating from min/max instead of /
Forgetting it represents 50% of data
Students often know a procedure related to mean absolute deviation (mad) but skip the recognition step: Is the spread found by averaging each value's plain distance from the mean, with no squaring? That leads to a calculation or graph that looks reasonable but answers a different question.
Forgetting absolute value
Dividing by wrong number
Confusing with standard deviation
Students often know a procedure related to standard deviation but skip the recognition step: Do you need the typical distance from the mean, found by squaring the distances, averaging, then taking a square root? That leads to a calculation or graph that looks reasonable but answers a different question.
Thinking SD can be negative (it can't)
Comparing SDs across different units
Confusing standard deviation (square root of variance) with variance itself
Students often know a procedure related to distribution shape but skip the recognition step: Are you describing the overall outline of a histogram or dot plot, such as one peak, two peaks, flat or lopsided? That leads to a calculation or graph that looks reasonable but answers a different question.
Confusing left/right skew direction
Expecting all data to be bell-shaped
Ignoring shape when choosing statistics
Students often know a procedure related to outlier detection but skip the recognition step: Must you decide whether particular values lie far enough from the rest to be flagged, by a fence or z-score cutoff? That leads to a calculation or graph that looks reasonable but answers a different question.
Automatically removing all outliers
Using only one detection method
Ignoring outliers' information
Students often know a procedure related to percentiles but skip the recognition step: Do you need the share of the group that falls below a value, stated as a rank out of 100? That leads to a calculation or graph that looks reasonable but answers a different question.
Confusing percentile with percentage score
Thinking 90th percentile means 90% correct
Interpolation errors
Students often know a procedure related to normal distribution but skip the recognition step: Are you modelling a bell-shaped variable fixed by its mean and standard deviation, and finding the probability of a range as an area? That leads to a calculation or graph that looks reasonable but answers a different question.
Assuming all data is normal
Confusing bell shape with exact normality
Forgetting the 68-95-99.7 rule
Students often know a procedure related to z-score (standard score) but skip the recognition step: Do you need to say how many standard deviations one value lies above or below its group's mean? That leads to a calculation or graph that looks reasonable but answers a different question.
Forgetting negative z-scores exist
Misinterpreting magnitude
Using with non-normal data carelessly
Students often know a procedure related to weighted average but skip the recognition step: Do some values count more than others, through percent weights, credits or group sizes? That leads to a calculation or graph that looks reasonable but answers a different question.
Forgetting to divide by the sum of weights
Using equal weights when data points have different importance
Confusing weights with the values themselves
Students often know a procedure related to empirical rule but skip the recognition step: Is the data roughly bell-shaped, with each boundary a whole number (1, 2 or 3) of standard deviations from the mean? That leads to a calculation or graph that looks reasonable but answers a different question.
Applying the rule to non-normal distributions
Confusing the percentages (e.g., saying 95% for one sigma)
Forgetting the rule gives approximate, not exact, percentages
Students often know a procedure related to skewness but skip the recognition step: Is the question about which way and how strongly a distribution's long tail stretches, away from its peak? That leads to a calculation or graph that looks reasonable but answers a different question.
Confusing the tail direction with where most data lies
Assuming all distributions are symmetric
Forgetting that outliers heavily influence skewness