Why do companies like to recruit median?

According to the Report on Enterprise Recruitment Salary in China, the national average monthly recruitment salary in the first quarter was 10 10 1 yuan. But it should be noted that this figure is only the average, not the median. Therefore, I think the median is more valuable than the average.

First of all, the median can better reflect the actual salary level. Different from the average, the median is the middle value arranged according to the salary from low to high, which is not affected by the extreme value in the sample. This means that the median can better reflect the income level of ordinary people in the workplace, rather than being affected by extremely high or low salaries.

Secondly, for the distribution of large sample data, the median is more accurate. To give an extreme example, if the CEO of a company earns as much as 1 10,000 yuan a month, while other employees earn only 5,000 yuan a month, then the average monthly salary will be much higher than the actual level. And if we use the median, we will see the actual income level of ordinary employees in this company more clearly.

Finally, the median stability is high. If the extreme value in the sample changes, the average value is easily affected and fluctuates greatly, while the change of the median value is relatively small. This means that the median is more suitable for long-term comparison, because they are less likely to be distorted by data emergencies.

Of course, the median also has certain limitations. For example, when the sample size is small, the median may not reflect the real situation well. In addition, the median can only reflect some basic trends and characteristics, but a more comprehensive analysis method is needed for complex data phenomena.

To sum up, although the average is one of the important statistics used to calculate the salary level, we should pay more attention to the reference value of the median. Only by comprehensively considering the influence of extreme value, the accuracy of big data samples and the stability of long-term comparison can we truly and comprehensively understand the actual situation of salary level.