skip to primary navigation skip to content
 

 

Drivers of entrepreneurship and small business

Note: WPs 1-22 refer to a Campop enhanced version of I-CeM (Campop I-CeM v.2.3). An updated version of I-CeM at UKDS will contain the enhancements used at Campop and for BBCE.

WP 1: Drivers of Entrepreneurship and Small Businesses: Project overview and outline of database construction

Bob Bennett, Harry Smith, Carry van Lieshout and Gill Newton, May 2017.

The census 1851-1911 provides information on employers and small firms that allows 200,000 - 2m individual entrepreneurs per year to be identified and the population of businesses to be estimated. This information has not been previously systematically available or researched. This paper outlines how information is extracted, how gaps and truncations in the I-CeM electronic records of the census are overcome, and how the data can be enriched. The paper outlines how entrepreneurship is defined from census and other records, how entrepreneurship rates can be calculated, and how a database has been created that is as complete and consistent as possible to be deposited at UKDA.

Downloadable working paper

WP2: Employers and the self-employed in the censuses 1851-1911: The census as a source for identifying SMEs, entrepreneurs, business numbers and size distribution

Bob Bennett, Harry Smith and Carry van Lieshout, May 2017.

This paper discusses how the original manuscript Census records can be used to derive information on employers and their workforce, self-employed sole proprietors, and the acreage of farms. It assesses how the content changes over time and has to be handled in database construction, focusing on issues of completeness; identification of employers, own account and workers; occupational coding; gender coverage; identification of portfolio businesses; partnerships; and location. This underpins database design for Entrepreneurs 1851-1911 to be deposited at UKDA.

Downloadable working paper

WP 3: Identifying businesses and entrepreneurs from the Censuses, 1851-1881

Carry van Lieshout , Bob Bennett, Harry Smith, and Gill Newton, May 2017.

This paper describes how individual entrepreneurs can be identified and extracted from the original Census Enumerators Books (CEBs) for 1851-81. It describes the early censuses and the method through which different groups of entrepreneurs can be identified, extracted and parsed, particularly focusing on identifying employee numbers of the businesses. It distinguishes between: employers with employees; own account proprietors; and people owning certain business assets. Constraints in the data are summarized and how information was infilled from other sources to correct for the substantial gaps and truncations in the I-CeM electronic records of the census.

Downloadable working paper

WP4: Identifying businesses and entrepreneurs from the Censuses, 1891-1911

Harry Smith, Bob Bennett, Carry van Lieshout and Gill Newton, May 2017.

This paper explains how employers and own account self-employed individuals can be extracted from the Census Enumerators Books (CEBs) for 1891-1911. The paper describes how they are identified, the issues arising, how miscoding in I-CeM is corrected, and how each census is cleaned and prepared for subsequent analysis. Detailed discussion covers quality of returns of business proprietor status, how far individuals had two or more statuses, the extent of biases in non- responses, and the geography of non-response. Use of occupation descriptors to clean the coding of employment status are steps towards database construction.

Downloadable working paper

WP 5: Business sectors, occupations and aggregations for business groups, 1851-1911

Bob Bennett, Harry Smith, Carry van Lieshout and Gill Newton, May 2017.

This paper discusses alternative aggregations that can be used to code business sectors using the census information on different occupations and employment status categories of entrepreneurs and workers. It defines two classifications at two levels of aggregation that are used in subsequent analysis: ID18 and ID50 used for sector/entrepreneurship coding; and EA17 and EA51 used for sector/whole economically active coding. It compares this coding with the more detailed coding in I-CeM, that used for the PST occupational classification system devised by Wrigley, and with the codes used in modern mainstream industrial classifications of sectors (SIC) and occupational groups. There is an accompanying data download of the classification file.

Downloadable working paper

Data download of classification file documented in WP 5

Download of Feinstein extension to sector classification

WP 6: Urban-Rural Classification using Census data, 1851-1911

Harry Smith and Bob Bennett, November 2017.

This paper develops an urban classification based on the census for 1851-1911 for England and Wales. The classification is based on the well-established Law-Robson urban classification but using the new data available from the e-version of the census. The paper discusses the Law-Robson criteria used to classify towns, and how urban parishes can be identified. The paper then presents the urban classification. An introduction to the documentation is given for possible data download.

Downloadable working paper

Downloadable classification E&W

Downloadable classification Scotland

Published paper in Urban History

WP 7: Classification of towns in 1891 using factor analysis

Harry Smith, Bob Bennett and Dragana Radicic, November 2017.

This paper discusses how census data for 1851-1911 can be used to classify the occupations of the economically active population within Registration Sub-Districts (RSDs) for England and Wales 1851-1911 using a factor analysis methodology applied to all occupations. This paper is a pilot for 1891 for all occupied population, and by gender.

Downloadable working paper

This paper forms the basis for an Urban History article.

WP 8: Classification of occupations for economically active: Factor analysis of Registration Sub-Districts (RSDs) in 1891

Bob Bennett, Harry Smith and Dragana Radicic, November 2017.

This paper discusses how factor analysis can be used to classify the towns and urban areas in England and Wales of over 10,000 population using the occupational structure of their populations derived from census data for 1851-1911. The urban classification derives from working paper 6. This paper gives a classification by occupation for1891 as a pilot year for all occupied population, and by gender, as well as by employment status as worker, employer or own account self-employed.

Downloadable working paper

Downloadable classification by RSD

WP 9: Reconstructing entrepreneur and business numbers for censuses 1851-81

Bob Bennett, Piero Montebruno, Harry Smith and Carry van Lieshout, February 2018, revised version March 2019.

This paper discusses how census data for 1851-1881 can be supplemented to give the employment status of individuals as own account self-employed, employers, or workers. The aim is to align information on entrepreneurs given in these earlier censuses with the information given in the later censuses from 1891 up to the present. The paper describes the supplementation methods used and provides the essential background documentation for using the data at aggregate and individual level. All individuals are coded in BBCE (EMPLOYSTATUS_NUM method); downloads give the decisions by Occode, file of all workers (from which all economically active derive - entrepreneurs are in BBCE database), and farm cut-offs for England and Wales: Scotland with WP 20

Downloadable working paper

Downloadable decisions for NUM method

Downloadable file of worker status 1851-1911 NUM method E&W

Data download of farm cut-offs

WP 9.2: Reconstructing business proprietor responses for censuses 1851-81: a tailored logit cut-off method

Bob Bennett, Piero Montebruno, Harry Smith and Carry van Lieshout, August 2019

This paper extends the supplementation in WP 9 to identify entrepreneurs 1851-81 at an individual level , where status was not explicitly identified in the censuses. The method developed uses a method of variable logit cut-offs tailored to each occupation code; all individuals are coded in BBCE (EMPLOYSTATUS_IND method); downloads here give the different decisions and workers for this method (farmers are unchanged) from which all economically occupies are derived. Scotland with WP 20.

Downloadable working paper

Downloadable decisions for IND method

Downloadable file of worker status 1851-1911 IND method E&W

WP 10: Classification of environments of entrepreneurship: Factor analysis of Registration Sub-Districts (RSDs) in 1891

Bob Bennett, Harry Smith and Dragana Radicic, August 2018.

This paper discusses how census data for 1851-1911 can be used to classify the employment status of the economically active population within Registration Sub-Districts (RSDs) for England and Wales using a factor analysis methodology based on 1891 as a pilot for other census years. Employment status in the 1891 census is given for (i) employers, (ii) those working on own account without employees, and (iii) workers. Codes are in BBCE.

Downloadable working paper

Downloadable classification by RSD

WP 11: Adjustment Weights 1891-1911: Weights to adjust entrepreneur numbers for non-response and misallocation bias in Censuses 1891-1911

Piero Montebruno, August 2018.

This paper explains the use of weights to adjust the Censuses 1891-1911 for non-response and misallocation bias. The weights are available for download (see below). The weights adjust observations to 'correct' values when using data from BBCE or I-CeM for the entrepreneurs identified for 1891-1911. The paper documents of how to adjust and weight the data using the data download.

Downloadable working paper

Download of updated adjustment weights for non-response bias 1891-1911 E&W

WP 12: Extraction of data on Entrepreneurs from the 1871 Census to supplement I-CeM

Carry van Lieshout, Joe Day, Piero Montebruno, and Bob Bennett, September 2018.

This paper describes how the database for entrepreneurs in the 1871 census was created and deposited for ESRC project ES/M010953. This project uses I-CeM as its main source; however, this does not cover England and Wales in 1871. This paper describes how data from an alternative supplier, S&N [theGenealogist.co.uk] was extracted and aligned with I-CeM to provide a full database for 1871 now in BBCE with additional codes as required.

Downloadable working paper

WP 13: Extracted data on employers and farmers compared with published tables in the Census General Reports, 1851-1881

Carry van Lieshout, Bob Bennett and Harry Smith, February 2019.

This paper compares published census tables for employers in England and Wales with extractions of individuals from the I-CeM and S&N sources now in BBCE (as variable GROUP). This is for non-farmers for 1851, and farmers for 1851-71, the only years that have published census tables for these data. The comparisons confirm a generally good match. Although there are some deficiencies possible from I-CeM/S&N data, the data are less ambiguous in definitions and more complete in some respects than achieved by published census tables, especially for larger firms and larger farms.

Downloadable working paper

WP 14: Company Directors: Directory and Census Record Linkage, 1881-1911

Carry van Lieshout, Bob Bennett, and Piero Montebruno, February 2019.

This paper describes how information on company directors has been identified in census records 1881-1911, and how this has been linked to directors listed in the Directory of Directors (DoD). Record linkage is used to match DoD to the census, achieving a 36% match rate overall for 18,200 directors. Linked directors in census are in the BBCE variable (DIR). The paper also defines the coding of business sectors, locations, and roles played by the directors in each company.

Downloadable working paper

WP 15: Entrepreneurial discrete choice: Modelling decisions between self-employment, employer and worker status

Bob Bennett, Piero Montebruno, Harry Smith, and Carry van Lieshout, February 2019.

A model between different forms of business organisation is estimated, focusing on the choice to operate as an entrepreneur as a sole proprietor on own account, or as an employer, rather than as a waged employee. The paper shows that marital status, gender, family structure, and spatial location are key factors behind differences between entrepreneurs and waged workers.

Downloadable working paper

WP 16: Constructing Parish-level Data and RSD-level Data on Transport Infrastructure in England and Wales 1851-1911

Max Satchell, Bob Bennett, Dan Bogart and Leigh Shaw-Taylor, February 2019.

This paper documents the methods used to construct a database at parish and RSD-level of transport infrastructure in England and Wales for each census year 1851-1911: for ports; turnpike roads (1851 only); first class roads (1911 only); railway lines and stations; waterways; and presence of sea coasts.

Downloadable working paper

WP 17: Building a 1911 Historical Land Capacity GIS

Max Satchell and Bob Bennett, April 2019

This paper outlines the construction of a historical land capacity GIS for census parishes and Registration Sub-Districts for c. 1911. It combines historical and modern land use and land capacity data and adjusts this to give an GIS for 1911.

Downloadable working paper

WP 18: Identification of business partnerships in the British population censuses 1851-1911 for BBCE

Bob Bennett, Harry Smith, Piero Montebruno and Carry van Lieshout, September 2019

This paper describes the process of identifying and coding business partners and partnerships from the population censuses 1851-1911 in England, Wales and Scotland derived from I-CeM to include in the British Business Census of Entrepreneurs (BBCE) data deposit.

Downloadable working paper

WP 19: Datasets and guide: downloads for reconstructing British census responses 1851-1881 for BBCE

Piero Montebruno, September 2019

This paper is a guide to the supplementation used in WP 9, WP 9.2 and WP 20 for the 1851-1881 censuses where responses were incomplete (some not explicitly identifying entrepreneur status). The paper provides downloads of the intermediate variables used in England and Wales, and Scotland.

Downloadable working paper

This paper has downloads of the reconstruction stages for the data for each year listed within WP in the same DOI. This includes files by RecID for all Scotland, and England and Wales for all economically active 1851-81: employers, own account, and workers.

WP 20: Preparing Scottish census data in I-CeM for the British Business Census of Entrepreneurs (BBCE)

Harry Smith, Carry van Lieshout, Piero Montebruno, and Bob Bennett, September 2019

BBCE data from I-CeM are enriched, corrected and infilled for those missing in I-CeM to identify each entrepreneur in Scotland from the censuses 1851-1901. BBCE and I-CeM are linked for each entrepreneur to give a database for Scotland aligned with that in England and Wales. All individuals are coded in BBCE (EMPLOYSTATUS_IND method); downloads give adjustment weights for non-response bias 1891-1901, supplementation decisions 1851-81, farmers by entrepreneur status, and RecID for all workers (from which all economically active derive); files for entrepreneurs are in BBCE database. E&W with WPs 9 and 9.2.

Downloadable working paper

Download of adjustment weights for non-response bias 1891-1901 Scotland

Downloadable decisions for IND method Scotland

Downloadable file of workers for IND method Scotland

Data download of farm cut-offs

WP 21: Additional codes and people in the British Business Census of Entrepreneurs (BBCE) not available through I-CeM

Carry van Lieshout, Robert J. Bennett, and Harry Smith October 2019

The BBCE includes many codes not in I-CeM which can be linked through the individual identifiers (RecID). This paper provides further small data corrections of I-CeM for sex, age and marital status; it also gives codes and downloads for additional census respondents omitted in FMP and I-CeM. Full data downloads for 1871 not in I-CeM are given in the BBCE.

Downloadable working paper

Data downloads

WP 22: Employers and self-employed in the censuses 1921-2011 and alignment with BBCE: Entrepreneurs, business numbers and size distribution

Bob Bennett, January 2020

How the published census reports from 1921 to 2011 can be used to derive information on employers and their workforce, self-employed sole proprietors, and workers. The paper discusses how these can be aligned with the earlier censuses to allow a continuous time series for 1851-2011, giving definitions and background to data used in BBCE and Bennett et al. (2019) 'The Age of Entrepreneurship: Business Proprietors, Self-employment and Corporations since 1851' (Routledge).

Downloadable working paper

WP 23: Missing in the Census 1851-1911: The 'lost', 'missing', and 'gaps' in I-CeM and BBCE, and weights to adjust RSD populations

Bob Bennett, Carry van Lieshout and Kevin Schürer, February 2020

WP 23 outlines the various gaps in e-census records from BBCE and I-CeM and how they have arisen. It assesses sources for identifying archivally lost records, those not in FMP and hence I-CeM, and compares with BBCE and I-CeM databases at the level of RSDs for census years 1851-61, and 1881-1911. The paper develops weights to compensate for the identified gaps and provides downloads of the weights for use with I-CeM and/or BBCE.

Downloadable working paper

Downloadable adjustment weights for missing data E&W by RSD 1851-1911

WP24: Record-linkage of entrepreneurs in the England and Wales Censuses 1851-91 using BBCE and I-CeM

Gill Newton and Bob Bennett, February 2020.

WP 24 uses a simple and efficient computational method for inter-census record linkage based on Jaro-Winkler, with fuzzy name matching and data blocking including birthplace and other standardisation. Entrepreneurs and a general population sample are derived from the BBCE and I-CeM England and Wales census datasets 1851-91. Matches are achieved with high precision (>0.9) for 10-30% of records in the candidate match pool, with recall rates of between >0.16 and >0.32.

Downloadable working paper

WP 25: Inter-census record-linked entrepreneurs and non-entrepreneurs 1851-91 using BBCE and I-CeM: database structure, assessment, downloads and User Guide

Piero Montebruno and Bob Bennett, February 2020.

WP 25 provides the database download, assessment of representativeness, and explanation of database finalisation for record-linkage of census data 1851-91 in England and Wales using the Jaro-Winkler method, with fuzzy name matching, data blocking, and data standardisation. The linked database can be downloaded; it is for two sets of individuals: entrepreneurs, and a stratified quota random sample on non-entrepreneurs in 21 occupational sectors.

Downloadable working paper

Downloadable database record-linked Entrepreneurs 1851-91

Downloadable database sample record-linked Non-entrepreneurs 1851-91

WP 26: Supplement to BBCE User Guide: Website definitions, downloads, Atlas of Entrepreneurship, and linkage to I-CeM

Bob Bennett, Carry van Lieshout, Harry, Smith and Piero Montebruno, March 2020.

WP 26 paper has two objectives: to guide users on the precise data definitions used for the BBCE website and for the Atlas of Entrepreneurship; and to offer additional guidance to that in the BBCE User Guide on how BBCE data should be linked with I-CeM. Downloads here also give resources additional to other WPs.

Downloadable User Guide Supplement

Additional data downloads to BBCE not listed under other WPs:

Additional codes for portfolio businesses (Group 1 stating employees) as used in J. Rural Studies (2017). The individuals are identified by GSU ID and occode as in Schürer, K. and Woollard, M. (University of Essex) (2000) 1881 Census for England and Wales, the Channel Islands and the Isle of Man (Enhanced Version) [computer file] UKDA, SN-4177:

Download of split workforces for each business with portfolios 1881

WP 27: Estimates of GB firm-size by sector in 1881 using BBCE sector definitions (EA17 sector codes)

Bob Bennett, Harry Smith and Les Hannah, October 2021.

This paper develops a methodology of data re-scaling of employer responses for number of employees in the censuses for 1851-81 to allow for non- response and biases in digital data capture. It uses BBCE EA17 sector codes to estimate firm-size by sector for 1881. It records work in progress, subject to revision of method and firm-size estimates.

Downloadable working paper