How much does the average word count differ between sectors?

Average word count and sample size

Figure 10: Average word count and sample size

Looking at the sample size of each sector is important to see the variation between them because some of them just look like that they are based on very small observations.

It’s important of being careful before making to any conclusions on a single graph that shows a single statistics of word count on all sectors. Here is a plot showing the average of word counts in sectors with their sample size, which is the number of narrative records in the group.

In that consideration, for text mining, it would be better to consider the sectors with which the average word count value is greater than and equal to 100. Therefore, those sectors are:

  • Poblacion (M = 34.61, SD = 55.40) with 204 sample size

  • Tecnico (M = 52.81, SD = 46.71) with 316 sample size

  • VBG (M = 55.00, SD = 48.35) with 132 sample size