Census data from the Languedoc-Roussillon
1954 - 1975

The data consist of the numbers of persons in 9 occupational categories (2nd mode) gathered during four censusses (3rd mode) in all 42 rural districts in the Languedoc-Roussillon, South-West France (1st mode). The data have been subjected to various analyses (see reference list)

The 3-way data array X = (x(i,j,k)) has the following form

          |-----------|i=1
     |-----------|    |i=2
|-----------|    |    |..
|           |    |    |..
|           |    |____|i=I=42    k=K=4
|           |____|             k=2
|___________|               k=1

  j=1,..,J=9

      

  j=1,..,J=9
|-----------|i=1
|           |i=2
|           |..     k= 1
|           |..
|___________|i=I=42
|-----------|i=1
|           |i=2
|           |..     k= 2
|           |..
|___________|i=I=42
|-----------|i=1
|           |i=2
|           |..     k=4
|           |..
|___________|i=I=42

      

Thus the first mode (i) is nested in the third mode (k) and the data consists of [42 (districts) rows times 4 (censusses)] rows and 9 (occupations) columns.

Lists and codes for the cantons, occupations, and years are contained in the following Word-documents:

• Cantons by year table of standarised residuals
• Cantons by occupation marginal frequencies.

Download the zipped data

References

• Primary reference
  Lavit, C. (1988).
  Analyse conjointe de tableaux quantitatifs [Simultaneous analysis of several quantitative matrices] Paris: Masson.
  Chapter 8 contains an analysis with STATIS; pp. 188-192 a complete listing of the data.
• Carlier, A. & Kroonenberg, P. M. (1998).
  The case of the French cantons: An application of three-way correspondence analysis. In J. Blasius & M. Greenacre (Eds.), Visualization of categorical data (pp. 253-275). New York: Academic Press.