If A Data Set Is Right Skewed, Where Would The Mean Be In Relation To The Mode?

Measures of Fundamental Tendency

Introduction

A measure of central trend is a single value that attempts to draw a set of data by identifying the fundamental position inside that set of data. As such, measures of central tendency are sometimes called measures of cardinal location. They are too classed as summary statistics. The mean (often called the average) is nigh likely the measure of central tendency that y'all are most familiar with, but at that place are others, such as the median and the fashion.

The mean, median and manner are all valid measures of cardinal tendency, but under different conditions, some measures of central tendency go more than appropriate to employ than others. In the following sections, we will look at the mean, mode and median, and learn how to calculate them and nether what conditions they are most appropriate to be used.

Mean (Arithmetics)

The hateful (or average) is the most popular and well known measure out of central trend. It tin can exist used with both discrete and continuous data, although its apply is most frequently with continuous data (see our Types of Variable guide for data types). The hateful is equal to the sum of all the values in the data set up divided by the number of values in the data set. And then, if we take $ n $ values in a information ready and they take values $ x_1, x_2, $ …$, x_n $, the sample mean, commonly denoted by $ \overline{x} $ (pronounced "x bar"), is:

$$ \overline{10} = {{x_1 + x_2 + \dots + x_n}\over{n}} $$

This formula is usually written in a slightly different manner using the Greek capitol letter of the alphabet, $ \sum $, pronounced "sigma", which ways "sum of...":

$$ \overline{x} = {{\sum{x}}\over{n}} $$

Yous may have noticed that the above formula refers to the sample hateful. So, why have we chosen it a sample mean? This is considering, in statistics, samples and populations take very different meanings and these differences are very important, even if, in the case of the hateful, they are calculated in the same way. To acknowledge that we are calculating the population mean and not the sample mean, we use the Greek lower case letter of the alphabet "mu", denoted as $ \mu $:

$$ \mu = {{\sum{x}}\over{n}} $$

The mean is essentially a model of your data fix. It is the value that is about common. You lot will observe, however, that the mean is not often 1 of the bodily values that you lot have observed in your information set. However, ane of its of import properties is that information technology minimises error in the prediction of any 1 value in your data ready. That is, it is the value that produces the lowest amount of fault from all other values in the data set.

An of import holding of the mean is that it includes every value in your information set every bit role of the adding. In addition, the mean is the simply mensurate of fundamental trend where the sum of the deviations of each value from the mean is always naught.

When not to utilise the hateful

The mean has ane main disadvantage: it is peculiarly susceptible to the influence of outliers. These are values that are unusual compared to the residuum of the information fix by being especially pocket-size or large in numerical value. For example, consider the wages of staff at a manufactory below:

Staff	1	2	3	4	5	6	7	8	nine	ten
Bacon	15k	18k	16k	14k	15k	15k	12k	17k	90k	95k

The mean salary for these x staff is $30.7k. However, inspecting the raw data suggests that this mean value might non be the best way to accurately reflect the typical salary of a worker, as almost workers have salaries in the $12k to 18k range. The mean is beingness skewed by the two large salaries. Therefore, in this situation, we would like to have a improve measure out of central tendency. As nosotros will find out afterwards, taking the median would be a ameliorate measure of central trend in this state of affairs.

Another time when we commonly prefer the median over the mean (or mode) is when our data is skewed (i.east., the frequency distribution for our data is skewed). If we consider the normal distribution - equally this is the almost frequently assessed in statistics - when the data is perfectly normal, the hateful, median and style are identical. Moreover, they all represent the about typical value in the data gear up. However, every bit the data becomes skewed the mean loses its ability to provide the best central location for the data because the skewed data is dragging it away from the typical value. All the same, the median best retains this position and is non as strongly influenced by the skewed values. This is explained in more particular in the skewed distribution section afterwards in this guide.

Median

The median is the middle score for a set of information that has been arranged in order of magnitude. The median is less afflicted past outliers and skewed data. In lodge to summate the median, suppose we accept the information below:

fourteen

We first need to rearrange that information into gild of magnitude (smallest first):

fourteen

Our median mark is the eye marker - in this case, 56 (highlighted in bold). It is the eye mark because there are 5 scores before it and 5 scores after it. This works fine when you take an odd number of scores, just what happens when you take an even number of scores? What if you had but x scores? Well, you simply accept to take the heart two scores and average the result. Then, if we await at the instance below:

We over again rearrange that data into guild of magnitude (smallest first):

xiv

But now we have to take the 5th and 6th score in our data set and average them to get a median of 55.5.

Mode

The mode is the most frequent score in our data set. On a histogram it represents the highest bar in a bar chart or histogram. You lot can, therefore, sometimes consider the style as being the almost popular pick. An instance of a mode is presented below:

Histogram showing mode as highest bar in the middle of the continuous distribution as the mode

Commonly, the fashion is used for categorical data where we wish to know which is the most common category, as illustrated below:

Bar chart showing highest bar as the mode

We can see to a higher place that the most common class of transport, in this particular data set up, is the bus. However, one of the problems with the mode is that information technology is non unique, so it leaves us with problems when we have two or more values that share the highest frequency, such every bit below:

Histogram of a continuous distribution showing two modes, both somewhat centrally located

We are now stuck every bit to which mode best describes the central tendency of the data. This is particularly problematic when nosotros have continuous data because we are more likely not to accept any one value that is more frequent than the other. For instance, consider measuring 30 peoples' weight (to the nearest 0.1 kg). How probable is it that we will find ii or more people with exactly the same weight (east.g., 67.four kg)? The answer, is probably very unlikely - many people might be close, simply with such a small sample (30 people) and a large range of possible weights, yous are unlikely to discover ii people with exactly the same weight; that is, to the nearest 0.1 kg. This is why the mode is very rarely used with continuous information.

Another problem with the mode is that information technology will non provide united states of america with a very good mensurate of central tendency when the most mutual marker is far abroad from the rest of the information in the information set, as depicted in the diagram below:

Histogram of a continuous distribution showing mode not centrally located

In the above diagram the mode has a value of 2. We tin clearly see, however, that the mode is not representative of the data, which is mostly concentrated around the xx to thirty value range. To use the mode to describe the central trend of this information set would be misleading.

Skewed Distributions and the Mean and Median

We often test whether our information is normally distributed because this is a common supposition underlying many statistical tests. An case of a ordinarily distributed set of information is presented below:

A histogram showing a normally distributed continuous data set

When y'all have a usually distributed sample you can legitimately use both the mean or the median as your measure of fundamental tendency. In fact, in any symmetrical distribution the mean, median and manner are equal. Notwithstanding, in this situation, the mean is widely preferred as the best measure of central trend because it is the measure that includes all the values in the data set up for its adding, and any change in any of the scores will affect the value of the hateful. This is not the example with the median or mode.

However, when our data is skewed, for example, as with the right-skewed data set below:

Histogram of a skewed distribution showing a noticable difference between the median and mean values

We find that the mean is being dragged in the straight of the skew. In these situations, the median is generally considered to be the best representative of the fundamental location of the data. The more skewed the distribution, the greater the divergence between the median and mean, and the greater accent should be placed on using the median as opposed to the mean. A archetype instance of the above right-skewed distribution is income (salary), where college-earners provide a fake representation of the typical income if expressed equally a mean and non a median.

If dealing with a normal distribution, and tests of normality show that the information is not-normal, it is customary to use the median instead of the mean. Even so, this is more than a rule of pollex than a strict guideline. Sometimes, researchers wish to report the mean of a skewed distribution if the median and mean are not appreciably different (a subjective assessment), and if it allows easier comparisons to previous research to exist made.

Summary of when to use the mean, median and mode

Please use the following summary table to know what the best measure out of central tendency is with respect to the dissimilar types of variable.

Type of Variable	Best measure out of central trend
Nominal	Mode
Ordinal	Median
Interval/Ratio (non skewed)	Mean
Interval/Ratio (skewed)	Median

For answers to ofttimes asked questions nearly measures of primal tendency, please go the adjacent folio.