An average is a single value that represents the center of a dataset. It summarizes a group of numbers into one number that is most representative of the whole. In Six Sigma, the average is one of the first statistics calculated in the Measure phase of DMAIC. It tells the team where the process is centered, which is the starting point for every capability, control chart, and hypothesis test that follows.
Average in Six Sigma
An average is a single representative value for a set of numbers, most commonly the arithmetic mean — the sum of all values divided by the count of values.
In Six Sigma’s Measure phase, the process average (written as X̄, pronounced “X-bar”) is the centerline of the X-bar control chart and the reference point for process capability calculations. Three types of average are used in statistics: the mean (most common in Six Sigma), the median (middle value), and the mode (most frequent value). Together these are called measures of central tendency.
Table of contents
Key Takeaways
- An average is a single value that represents the center of a dataset. In everyday use, “average” most commonly means the arithmetic mean.
- The arithmetic mean is calculated as: Mean = Sum of all values / Count of values. It is the most widely used average in Six Sigma because it uses every data point in the calculation.
- Three types of average — mean, median, and mode — are each called measures of central tendency. Each one describes where a dataset is centered, but each uses a different method.
- In Six Sigma, the sample mean is written as X̄ (X-bar). The population mean is written as µ (mu).
- The X-bar chart in Statistical Process Control plots the average of subgroups over time. Its centerline is X̄-bar — the average of the subgroup averages.
- The arithmetic mean is sensitive to outliers. A single extreme value pulls the mean toward it. The median is resistant to outliers and better represents the center of a skewed dataset.
- The weighted average assigns different importance (weights) to different values. Six Sigma practitioners use it when data points do not contribute equally to the overall figure.
- The moving average smooths time-series data to reveal trends. Walter Shewhart’s control chart work, which forms the basis of Six Sigma SPC, relied on moving average concepts to detect process shifts.
What Is an Average?
An average is a number that stands in for an entire group of numbers. It is a summary — a single value chosen to represent where the group is centered. When someone asks “what is the average defect rate?” or “what is the average cycle time?”, they want one number that captures the typical or central value across many measurements.
The word “average” has a specific technical definition in statistics, but it also has a broader everyday use. In everyday language, people use “average” to mean any representative summary. In statistics, “average” refers specifically to a class of measures called measures of central tendency. The three measures in this class — mean, median, and mode — each represent the center of a dataset using a different calculation.
In Six Sigma, the arithmetic mean is the standard measure called “the average.” It uses every data point, responds to every change in the data, and connects directly to the statistical tools used across the Measure, Analyze, and Control phases of DMAIC.
Public, Onsite, Virtual, and Online Six Sigma Certification Training!
- We are accredited by the IASSC.
- Live Public Training at 52 Sites.
- Live Virtual Training.
- Onsite Training (at your organization).
- Interactive Online (self-paced) training,
The Four Types of Average
Four distinct types of average appear in statistics and in Six Sigma practice. Each suits a different data structure and analytical purpose.
1. Arithmetic Mean
The arithmetic mean is the most common type of average. It is calculated by adding all values in a dataset and dividing by the count of values.
Formula: X̄ = (x₁ + x₂ + x₃ + … + xₙ) / n
Where x₁ through xₙ are the individual values and n is the count of values.
Worked example: A team measures the cycle time (in minutes) for six transactions: 12, 15, 11, 18, 14, 16.
Sum = 12 + 15 + 11 + 18 + 14 + 16 = 86 Count = 6 Mean = 86 / 6 = 14.33 minutes
The arithmetic mean of 14.33 minutes represents the central tendency of this cycle time data. It tells the team that on average, each transaction takes approximately 14 and one-third minutes.
When to use it: The arithmetic mean works best for symmetric datasets without extreme outliers. It uses every data point, which means one very large or very small value pulls the mean toward it. In Six Sigma, the mean is used in process capability calculations, control chart centerlines, and hypothesis tests.
Six Sigma notation: The sample mean is written as X̄ (X-bar). The true population mean, rarely known in practice, is written as µ (mu). Statistical inference in Six Sigma estimates µ from X̄ using sampled process data.
2. Median
The median is the middle value in an ordered dataset. When values are arranged from smallest to largest, the median is the value exactly in the middle. For an even number of values, the median is the arithmetic mean of the two middle values.
Example with six values (even count): Ordered cycle times: 11, 12, 14, 15, 16, 18. The two middle values are 14 and 15. Median = (14 + 15) / 2 = 14.5 minutes
The median is resistant to outliers. If one transaction took 120 minutes instead of 18 minutes, the median would barely change. The mean, however, would jump significantly. This makes the median the better measure of central tendency for skewed data — data with a long tail in one direction.
Six Sigma context: The median appears in non-parametric hypothesis tests, such as the Mann-Whitney test and the Kruskal-Wallis test. These tests compare medians rather than means because they make no assumption about the distribution of the data. When process data fails the normality test, non-parametric methods that use the median become relevant in the Analyze phase.
3. Mode
The mode is the value that appears most often in a dataset. A dataset can have one mode (unimodal), two modes (bimodal), or more.
Example: Defect counts per shift over ten shifts: 3, 5, 3, 7, 3, 8, 5, 3, 9, 4. The value 3 appears four times — more than any other value. Mode = 3
Six Sigma context: The mode is less commonly used in quantitative Six Sigma analysis but appears when analyzing categorical data or when identifying the most frequently occurring defect type. A bimodal distribution — one with two modes — is an important diagnostic signal in the Measure phase. It indicates that two different sub-populations or process conditions have been mixed into one dataset without stratification.
4. Weighted Average
The weighted average assigns a different weight (degree of importance) to each value before averaging. Values with higher weights contribute more to the final average than values with lower weights.
Formula: Weighted Average = Σ(value × weight) / Σ(weights)
Example: A quality team scores three suppliers on delivery performance (scored 1-10) with different weights:
| Supplier | Score | Weight |
| Supplier A | 8 | 50% |
| Supplier B | 6 | 30% |
| Supplier C | 9 | 20% |
Weighted Average = (8 × 0.50) + (6 × 0.30) + (9 × 0.20) = 4.0 + 1.8 + 1.8 = 7.6
The unweighted mean of those scores would be (8 + 6 + 9) / 3 = 7.67. The weighted average adjusts the result to reflect that Supplier A handles 50% of deliveries and therefore contributes most to overall performance.
Six Sigma context: Weighted averages appear when calculating process performance metrics across production lines that run at different volumes, when aggregating supplier scorecards, or when combining data from different time periods with different sample sizes.
Mean vs. Median vs. Mode: When to Use Each
Choosing the right type of average depends on the data structure and the question being asked. This table summarizes the key differences:
| Feature | Arithmetic Mean | Median | Mode |
| Calculation | Sum ÷ count | Middle value in ordered data | Most frequent value |
| Uses all data points | Yes | No | No |
| Sensitive to outliers | Yes | No | No |
| Best for | Symmetric, normal data | Skewed data or data with outliers | Categorical data or identifying common patterns |
| Six Sigma primary use | Process capability, control charts, t-tests | Non-parametric tests, median charts | Bimodal detection, defect type analysis |
| Distribution assumption | Assumes rough normality for valid inference | None | None |
The arithmetic mean, median, and mode are equal in a perfectly symmetric, normal distribution. In a right-skewed distribution, the mean is greater than the median, which is greater than the mode. In a left-skewed distribution, the order reverses. Recognizing this relationship helps Six Sigma practitioners interpret histograms and understand whether the mean accurately represents the center of their process data.
Also Read: Arithmetic Mean: Guide to Understanding and Calculating Averages
The Moving Average and Its Role in Process Monitoring
A moving average (also called a rolling average or running average) calculates the mean of a rolling window of consecutive data points. As each new data point is added, the oldest one is dropped, and the average is recalculated.
Example — 3-point moving average: Daily defect counts: 8, 12, 10, 15, 9, 11, 14
| Day | Count | 3-Point Moving Average |
| 1 | 8 | — |
| 2 | 12 | — |
| 3 | 10 | (8+12+10)/3 = 10.0 |
| 4 | 15 | (12+10+15)/3 = 12.3 |
| 5 | 9 | (10+15+9)/3 = 11.3 |
| 6 | 11 | (15+9+11)/3 = 11.7 |
| 7 | 14 | (9+11+14)/3 = 11.3 |
The moving average smooths out day-to-day fluctuation and reveals whether the underlying trend is rising, falling, or stable. Random spikes (like day 4’s count of 15) are absorbed into the moving average rather than appearing as sharp peaks.
Six Sigma context: Moving averages appear in the EWMA (Exponentially Weighted Moving Average) control chart, which gives more weight to recent data points.
How the Average Is Used Across DMAIC
The average is not just a Measure phase concept. It connects to analytical decisions in every DMAIC phase.
| DMAIC Phase | Role of the Average |
| Define | The project Y is often expressed as an average: “reduce average cycle time from 14 minutes to below 10 minutes.” The average gives the problem statement a measurable starting point. |
| Measure | The sample mean (X̄) is calculated from baseline data and becomes the centerline of the X-bar control chart. Process capability indices Cp and Cpk use the sample mean to assess centering within specification limits. |
| Analyze | Hypothesis tests compare means between groups. The two-sample t-test tests whether the average differs between two groups (such as two machines or two shifts). One-way ANOVA tests whether the average differs across three or more groups simultaneously. |
| Improve | After implementing a change, the team recalculates the mean from the improved process and compares it to the baseline mean. A statistically significant shift in the mean confirms that the improvement worked. |
| Control | The X-bar chart plots subgroup averages over time and signals when the process mean has shifted. The centerline of the X-bar chart is X̄-bar — the average of the subgroup means — calculated from the baseline period. |
Also Read: Performance Baseline Measure: What It Is and How to Build One
The X-bar Chart: The Average in Action
The X-bar chart is the most direct application of the average concept in Six Sigma Statistical Process Control. According to the International Six Sigma Institute, when data is continuous, teams measure both the average and the variation, using an X-bar and R (Range) chart to monitor process performance.
The X-bar chart works as follows:
- Collect measurements in subgroups of a fixed size (typically 3 to 5 units per subgroup).
- Calculate the arithmetic mean (X̄) of each subgroup.
- Plot each subgroup mean on the chart over time.
- Calculate the centerline as X̄-bar (the average of all the subgroup means).
- Calculate the Upper Control Limit (UCL) and Lower Control Limit (LCL) at ±3 standard deviations from X̄-bar.
The X-bar chart monitors the central tendency of the process over time. When the process mean shifts — because a machine setting has drifted, a new operator uses a different technique, or a new material lot has a different property — the subgroup averages start clustering toward one of the control limits. The chart detects this shift and signals that a special cause requires investigation.
Frequently Asked Questions: Average
Q: What is an average in statistics?
A: An average is a single value that represents the center of a dataset. In statistics, three types of average are called measures of central tendency: the arithmetic mean (sum of values divided by count), the median (middle value in ordered data), and the mode (most frequently occurring value). In everyday use and in Six Sigma, “average” most commonly refers to the arithmetic mean.
Q: What is the formula for the arithmetic mean?
A: The arithmetic mean is calculated as: Mean = Sum of all values / Count of values. In Six Sigma notation, the sample mean is written as X̄ (X-bar). If six cycle times are 12, 15, 11, 18, 14, and 16 minutes, the arithmetic mean is (12+15+11+18+14+16) / 6 = 86 / 6 = 14.33 minutes.
Q: What is the difference between mean, median, and mode?
A: The mean is the arithmetic average — all values summed and divided by the count. The median is the middle value in ordered data. The mode is the most frequently occurring value. In a symmetric, normal distribution all three are equal. In a skewed distribution they diverge. The mean is sensitive to outliers; the median is not.
Six Sigma uses the mean for most parametric analysis and the median for non-parametric tests on skewed or non-normal data.
Q: How is the average used in Six Sigma’s DMAIC framework?
A: The average is used throughout DMAIC. In the Define phase, the project Y is often expressed as a target average. In the Measure phase, the sample mean becomes the centerline of the X-bar control chart and is used in process capability calculations. Further, in the Analyze phase, hypothesis tests compare means between groups to identify statistically significant differences.
While in the Improve phase, the team compares the post-improvement mean to the baseline mean. In the Control phase, the X-bar chart monitors subgroup averages to detect mean shifts.
Q: What is a weighted average and when is it used in Six Sigma?
A: A weighted average assigns different levels of importance (weights) to different values before averaging. It is calculated as the sum of (value × weight) divided by the sum of weights. In Six Sigma, weighted averages are used when data points do not contribute equally to the overall metric — for example, when aggregating supplier performance scores across vendors with different delivery volumes, or when combining performance data across production lines that run at different rates.
Q: What is a moving average?
A: A moving average calculates the mean of a rolling window of consecutive data points. As each new data point arrives, the oldest one is dropped and the average is recalculated.
Moving averages smooth out short-term fluctuations in time-series data to reveal underlying trends. In Six Sigma’s Control phase, the EWMA (Exponentially Weighted Moving Average) chart builds on this concept and is particularly useful for detecting small, sustained shifts in the process mean.
Q: Why does Six Sigma use X-bar notation for the average?
A: X-bar (X̄) is the standard statistical notation for a sample mean. The bar over the X indicates that the values have been averaged. In control charts, X̄-bar (with a bar over the entire X̄ symbol) represents the average of the subgroup means — the grand mean of the data collected during the baseline period.
This notation distinguishes the sample mean (X̄) from the true but unknown population mean (µ) that Six Sigma practitioners are trying to estimate and control.
Average Training in Six Sigma
The arithmetic mean, median, mode, weighted average, and moving average are foundational concepts introduced at the Yellow Belt level and applied throughout the Green Belt and Black Belt curriculum. Understanding which type of average to use in a given situation — and understanding why the mean can mislead when data is skewed or contains outliers — is a practical skill that shapes every data analysis decision in a DMAIC project.
At Six Sigma Development Solutions, measures of central tendency including the average are taught in context across our Yellow Belt, Green Belt, and Black Belt training programs. Practitioners learn not just the formula, but how to choose the right measure, how to spot when the mean does not accurately represent the data, and how the average connects to control charts, hypothesis tests, and capability analysis.
We offer training in three formats:
- Onsite training — delivered at your facility, using your actual process data in statistical exercises.
- Live virtual training — instructor-led sessions online with real-time interaction and Minitab exercises covering means, control charts, and capability analysis.
- Online training — self-paced Yellow Belt, Green Belt, and Black Belt certification programs covering all IASSC-testable central tendency content.
Explore our Six Sigma training programs or contact our team to find the right program for your goals.
About Six Sigma Development Solutions, Inc.
Six Sigma Development Solutions, Inc. offers onsite, public, and virtual Lean Six Sigma certification training. We are an Accredited Training Organization by the IASSC (International Association of Six Sigma Certification). We offer Lean Six Sigma Green Belt, Black Belt, and Yellow Belt, as well as LEAN certifications.
Book a Call and Let us know how we can help meet your training needs.
