Detailed 1931 occupational statistics for counties and large towns

Table ID:
OCC_1931_LG_T     (6566498)
Contents:
Detailed 1931 occupational statistics for counties and large towns
Approx. number of rows:
159,064
Table type:
Raw Data
Documentation Author:
Humphrey Southall
Geography:
Reporting units are identified by:
   Local Government Unit
   Local Government District Type
   Administrative County
Chronology:
The data cover the period 1881 to 1911.

Sources:

  1. These data are a full transcription of Table 16, 'Occupations of Males and Females aged 14 years and over showing also the total "Operatives" and the total "Out of Work" for England and Wales, Regions, Administrative Counties, County Boroughs and other Urban Areas with Populations exceeding 50,000', pp. 154-425 in: Census, 1931: Classification of Occupations, (London: HMSO, 1934).


Notes:

  1. Each place has many rows of data, one for each summary occupation, occupational sub-group or specific occupation. Male and female data are recorded on the same row.
  2. The original table has two columns for each unit covered, one for males and one for females, with the occupational categories in rows and extending over several pages. The data are held here, in a structure identical to that used for the equivalent 1951 data, with a separate row for each combination of reporting unit and occupational category.
  3. Although this dataset was originally input by CDDA, it was very extensively modified and much further error checking done by Andy Bendall and Humphrey Southall at Portsmouth in November 2000 and March 2001. The new error checking used Oracle to carry out additional column-wise check sums, partly to identify errors introduced when the data were re-organised in Belfast.


Checking:

  1. The total population was checked against the sum of those over and under 14.
  2. The population aged 14 and over was checked against the sum of those occupied and unoccupied.
  3. The percentage of operatives was compared with a computed percentage; NB this involves a division and therefore does not provide an absolute check on the precise number of operatives.
  4. The percentage of unemployed was compared with a computed percentage; NB this involves a division and therefore does not provide an absolute check on the precise number of unemployed.
  5. A computed total for all 32 occupational orders was compared with the reported population aged 14 and over; this is mainly a check that all occupational orders are present.
  6. A computed total of all the individual categories within each order, excluding any intermediate totals, was compared with the reported total for each order.
  7. A computed total of all the individual categories within each sub-order (i.e. level 2 category), again excluding any intermediate totals (see next note), was compared with the reported total for each sub-order. Corrections were made by using the computed total, not by checking against the original, as the component figures had already been corrected. NB not every order is divided into sub-orders.
  8. A computed total for each of the three ad hoc sub-totals was compared with the reported sub-total. These ad hoc sub-totals were for: 'Agricultural Labourers, Farm Servants' (row ID=22), 'Proprietors and Managers of Retail Businesses for the sale of' (row ID=488), and 'Salesmen and Shop Assistants retailing' (row ID=511). Corrections were made by using the computed total, not by checking against the original, as the component figures had already been corrected.
  9. A very small number of rows failed these checks because of unresolvable errors in the originals:
    • The reported male total for order 5 (Brick makers, etc) for Oldham, Lancashire, is greater than the computed total by 1, and the problem is within sub-order 1 (Makers of Bricks, Tiles and Pottery) for which the sub-total is also too large by 1.
    • The reported male total for order 22 (Transport) for Acton, Middlesex is greater than the computed total by 9, and the problem is within sub-order 4 (Other workers in transport and communications) for which the sub-total is also too large by 9.
    • The reported female total for order 27 (Persons engaged in Personal Service) for West Sussex Administrative County is smaller than the computed total by 1.
    • The reported male total for order 27 (Persons engaged in Personal Service) for Isle of Wight Administrative County is larger than the computed total by 1.


Indices:

IndexTypeColumn(s) indexed
occ_1931_lg_t_pkey Primary key lg_unit, lg_type, adm_cnty, row_id


Constraints:

The table has the following associated constraints:

ConstraintTypeDetails
occ_1931_lg_t_pkey Primary Key See details above for primary key index



Columns within table:

ColumnTypeContents
lg_unit Text string (max.len.=44). Name of the unit the data relates to. This may be a local government district, but the table includes county totals, with and without any County Boroughs listed with the county.
lg_type Text string (max.len.=8). Type of district. This is limited to: 'AC' = Administrative County, 'ACB' = Administrative County with associated County Boroughs, 'CB' = County Borough, 'CC' = County Corporate (this is specific to the City of London) 'LB' = London Borough, 'MB' = Municipal Borough, 'UD' = Urban District.
adm_cnty Text string (max.len.=28). Name of administrative county in which the place is located.
row_id Integer number. Numeric field identifying and placing in order the different rows of information for each reporting unit; for example, 'Boiler firemen, stokers' (category 915, order XXV) will always have row_id=359.
lev_1 Floating point number. Level 1 occupational category, given in roman numerals in the original report. This column is non-null for all rows apart from the initial 'occupational status' data.
lev_2 Floating point number. Level 2 occupational category. This does not appear for all level 1 categories.
lev_3 Floating point number. Level 3 (lowest) occupational category; a three digit occupational code number; which is null where the row is a higher level total. NB in the original report, 'Retired from Previous Gainful Occupation' had the code 'number' 'Y' and 'Other Persons (no Gainful Occupation stated)' had the code 'X'. These have been coded as 998 and 999 respectively so that this could be a numeric column. Three rows provided ad hoc totals within categories, had no occupational code in the original report but were coded as 1000 so that they could be distinguished from both level 2 totals and figures for individual occupations. These were: 'Agricultural Labourers, Farm Servants' (row ID=22), 'Proprietors and Managers of Retail Businesses for the sale of' (row ID=488), and 'Salesmen and Shop Assistants retailing' (row ID=511).
occup Text string (max.len.=124). Name of the occupational category. NB this may be the name of a level 1, 2 or 3 category, or an occupational status. In most cases, it is simply whatever text appears next to the row of numbers.
males Floating point number. Number of males.
females Floating point number. Number of females.
ac_tot Text string (max.len.=6). Flag to indicate rows which are totals for Administrative Counties. This is always set to 'T' for rows where lg_type = 'ACB', and manually set to 'T' for rows where lg_type = 'AC' and there are no associated County Boroughs.
notes Long text. Notes.
rec_num Integer number. Sequence number added on loading, placing the data rows in an overall order.
admc_unit Integer number. ID number for the county containing the unit, as defined in the AUO. If the unit is a county, this will be identical to g_unit.
g_unit Integer number. ID number for reporting area, as defined in the AUO.