Histogram

From CEOpedia | Management online
Histogram
See also

Histogram is a compilation of data, in the form of a statistical chart consisting of adjacent bars (rectangles), which illustrates the number of occurrence of test features in population or sample.

The base (which rests on the horizontal axis) are size of class compartments.

This method of construction is used when a number of distribution ranges are equal. If the series has unequal intervals, the height of the rectangles is determined by the index of abundance (frequency) corresponding to the various classes.

Indices of abundance (frequency) shall be determined as follows\[{\text{IofA}} = \frac{k*i}{l}\]

where:

k - abundance of the class,
i - narrowest class interval
l - class interval.

The histograms are used mainly to present the structure of the population (or phenomenon), hence structural series of qualitative and quantitative characteristics.

Charts highlighting the basic features of the population (or phenomenon) must be precisely adapted to the nature of these features.

Histogram Construction

The Histogram as other statistical charts consists of several parts:

  • field,
  • bars,
  • scale,
  • title,
  • legend,
  • source.

The basis for drawing up a histogram that describes the accuracy of the occurring in population (phenomena) is a rectangular coordinate system, with the main attention to the selection of the scale and precise graphical image.

A special form of the histogram is cumulative histogram. The horizontal axis in a rectangular coordinate system the accumulated numbers are shown.

Benefits and limitations of histogram

Benefits of histograms include:

  • They provide a clear and intuitive visual representation of the distribution of a dataset.
  • They can reveal the presence of outliers and skewness in the data.
  • They can be used to identify potential errors in data collection or measurement.

Limitations of histograms include:

  • They can be affected by the choice of bin size and location, which can obscure important features of the data.
  • They are not well-suited for displaying continuous data or data with small sample sizes.
  • They do not provide information about the correlation between variables.
  • They can be misleading if the sample size is small or if the data is not representative of the population.

References