1 The seminal text P. Laslett ed., with the assistance of R. Wall, Household and family in past time: comparative studies in the size and structure of the domestic group over the last three centuries in England, France, Serbia, Japan and colonial North America, with further material from Western Europe (Cambridge, 1972) resulted largely from a conference called by Laslett and held in Cambridge in 1969.

2 However, it is also the case that relatively little has been published on this subject for England and Wales in the past decade.

3 Anderson, M., ‘Households, families and individuals: some preliminary results from the national sample from the 1851 census of Great Britain’, Continuity and Change 3, 3 (1988), 421–38CrossRefGoogle Scholar, here 421.

4 K. Schürer and E. Higgs, Integrated Census Microdata (I-CeM); 1851–1911 (computer file). Colchester, Essex: UK Data Archive (distributor), April 2014. SN: 7481, http://dx.doi.org/10.5255/UKDA-SN-7481-1. A user guide and manual to the I-CeM data is available as Higgs, E., Schürer, C. Jones, K. and Wilkinson, A., The Integrated Census Microdata (I-CeM) guide (Colchester, 2013)Google Scholar. Further details on the I-CeM database together with a number of related resources are available from the I-CeM website at https://www.essex.ac.uk/history/research/icem/. The creation of the I-CeM database was made possible through funding from the UK Economic and Social Research Council (ESRC), grant number RES-062-23-1629. The version of the I-CeM data used here has been enhanced as the result of work by Schürer, Garrett, Jaadla and Reid as part of the ESRC-funded ‘An Atlas of Victorian Fertility Decline’ project (ES/L015463/1) at the Cambridge Group for the History of Population and Social Structure, Department of Geography, University of Cambridge, and will be deposited with the UK Data Archive at a future date. For further details, see http://www.geog.cam.ac.uk/research/projects/victorianfertilitydecline/. It should be noted that the original version of the I-CeM data does not include the 1871 census of England and Wales. A version of the 1871 data has subsequently been made available to the authors of this article, however, it does not include transcribed data from marital status, occupation or place of birth. Therefore, unfortunately, 1871 has had to be excluded from most of the analyses presented in this article. The original version of I-CeM also includes a number of duplicated person records, especially for 1851 and 1861. Every attempt has been taken to delete these in the calculations reported in this article.

5 Anderson, ‘Households, families and individuals’, 423.

6 Higgs et al., I-CeM guide, 163–5, 289–308. A CFU consists of either a couple living together, or parent(s) with never-married child(ren). No individual can be in more than one CFU.

7 Since the creation of version 1 of the I-CeM data, a number of corrections and enhancements have been made to the database, particularly in relation to households that for various reasons had become split in the initial version. This and other problems, which affected a small proportion of households and their CFU allocation, have now been fixed. See endnote 3.

8 Higgs et al., I-CeM guide, 277–9.

9 All the censuses were de facto so there was no confusion over recording of absent or temporarily present household members.

10 General Report to the 1851 Census, xxxiv.

11 1891 General Report, v. The definition in 1891 omitted the term lodger altogether: ‘As a general rule, the term “occupier” is to be understood to apply to the resident owner, or to a person who pays rent whether for the whole of a house, or for a tenement consisting of one or more rooms.’ See http://www.histpop.org/ohpr/servlet/PageBrowser?path=Browse/Census%20(by%20date)/1891/England&active=yes&mno=61&tocstate=expandnew&display=sections&display=tables&display=pagetitles&pageseq=5.

12 Higgs, E., Making sense of the census: the manuscript returns for England and Wales, 1801–1901 (London, 1989), 58–62Google Scholar; Higgs, E., ‘Structuring the past: the occupational, social and household classification of census data’, Computing and History Today 4 (1988), 24–30Google Scholar; Schürer, K. and Mills, D. R., ‘Family and household structure’, in Mills, D. and Schürer, K. eds., Local communities in the Victorian Census Enumerators’ Books (Oxford, 1996), 282–4Google Scholar.

13 Such as number of children per family (for example, Census of England and Wales, 1861, Volume III: General report, 10–11). It was not until the 1951 census that a serious attempt was made to analyse the composition of private households. See Office of Population Censuses and Surveys and the General Register Office, Edinburgh, Guide to Census Reports, Great Britain 1801–1966 (London, 1977), 145–7Google Scholar.

14 This is consistent with the suggestions made in Anderson, M., ‘Standard tabulation procedures for the census enumerators’ books, 1851–91’, in Wrigley, E. A. ed., Nineteenth-century society: essays in the use of quantitative methods for the study of social data (Cambridge, 1972), 134–45CrossRefGoogle Scholar.

15 See, as an illustration, I. C. Taylor, ‘Liverpool's institutional and quasi-institutional populations in 1841 and 1851’, in Mills and Schürer eds., Local communities, 42–6.

16 This exercise was undertaken using a combination of address information and the ratio between familial and non-familial members. See Higgs, Jones, Schürer and Wilkinson, I-CeM Guide, for details.

17 P. Laslett, ‘Introduction’, in Laslett and Wall, Household and family in past time, 1–89, esp. 28–32.

18 See, especially, Cooper, D. and Donald, M., and, ‘Households “hidden” kin in early nineteenth-century England: four case studies in suburban Exeter, 1821–1861’, Continuity and Change 10, 2 (1995), 257–78CrossRefGoogle Scholar. For a discussion and illustration of the problem, see E. Higgs, ‘The tabulation of occupations in the nineteenth-century census, with special reference to domestic servants’, in Mills and Schürer eds., Local communities, 27–35.

19 The figures for the 1890–1899 Life Table presented in Table 2 for 1891 are artificially lowered due to high mortality levels in that year. The effect can also be seen in the 6^th English Life Table, which was calculated for the period 1891–1900, but is less marked. See Woods, R. and Shelton, N., An atlas of Victorian mortality (Liverpool, 1997), 27Google Scholar.

20 See also Reid, A. M., Garrett, E., Dibben, C. and Williamson, L. O., ‘Gender specific mortality trends over the epidemiological transition: a view from the British mainland 1850–2000’, in Dinges, M. and Weigl, A. eds., Gender-specific life expectancy in Europe, 1850–2010 (Stuttgart, 2016), 73–88Google Scholar.

21 There is a large literature on post-1850 mortality decline in Britain, see especially, Jaadla, H. and Reid, A. M., ‘The geography of early childhood mortality in England and Wales, 1881‒1911’, Demographic Research 37 (2017), 1861–90CrossRefGoogle Scholar; Woods, R., The demography of Victorian England and Wales (Cambridge, 2000)CrossRefGoogle Scholar; Garrett, E., Reid, A., Schürer, K. and Szreter, S., Changing family size in England and Wales: place, class and demography, 1891–1911 (Cambridge, 2001)CrossRefGoogle Scholar; Garrett, E., Galley, C., Shelton, N. and Woods, R. eds., Infant mortality: a continuing social problem (Aldershot, 2006)Google Scholar; Woods, R., ‘The effects of population redistribution on the level of mortality in nineteenth-century England and Wales’, Journal of Economic History 45, 3 (1985), 645–51CrossRefGoogle ScholarPubMed.

22 The SMAMs have been calculated from the individual level I-CeM data for England and Wales as part of the ESRC-funded ‘An Atlas of Victorian Fertility Decline’ project (see endnote 3). Note that SMAMs for England (less Monmouth) are given for the census years 1851–1911 in Wrigley, E. A. and Schofield, R. S., The population history of England, 1541–1871: a reconstruction (London, 1981), 437Google Scholar, Table 10. 3. These are consistently older in the case of both sexes, particularly so at the start of the period (M: 26.9; F: 25.8) suggesting that overall marriage occurred at younger ages in Wales compared to England. However, as the maps generated by the ‘An Atlas of Victorian Fertility Decline’ project illustrate, SMAMs in Wales displayed marked geographical variation, with the coalfields of south Wales being characterised by younger SMAMs and much of central and northern rural Wales having late SMAMs.

23 Wrigley and Schofield, Population history, 437.

24 See Garrett et al., Changing family size; Woods, Demography of Victorian England; Woods, R. I. and Smith, C. W., ‘The decline of marital fertility in the late nineteenth century: the case of England and Wales’, Population Studies 37, 2 (1983), 207–25CrossRefGoogle ScholarPubMed; Woods, R. I. and Hinde, P. R. A., ‘Nuptiality and age at marriage in nineteenth-century England’, Journal of Family History 10, 2 (1985), 119–44CrossRefGoogle ScholarPubMed; Teitelbaum, M. S., The British fertility decline: demographic transition in the crucible of the Industrial Revolution (Princeton, 1984)CrossRefGoogle Scholar.

25 See, especially, Woods, The demography of Victorian England and Wales; Woods and Shelton, Atlas of Victorian mortality; Garrett et al., Changing family size; Reid, A., ‘Locality or class? Spatial or social differentials in infant and child mortality in England and Wales 1895–1911’, in Corsini, C. A. and Viazzo, P. P. eds., The decline of infant and child mortality: the European experience: 1750–1990 (The Hague, 1997), 129–54Google Scholar.

26 As Ruggles has very poignantly put it: ‘In order to live with extended relatives, one must have extended relatives’; Ruggles, S., Prolonged connections: the rise of the extended family in nineteenth-century England and America (Wisconsin, 1987), 60Google Scholar. For the classic statement regarding the relationship between demography and household structure as captured by the census, see Burch, T. K., ‘The size and structure of families: a comparative analysis of census data’, American Sociological Review 32, 3 (1967), 347–63CrossRefGoogle Scholar; Berkner, L., ‘The stem family and the developmental cycle of the peasant household: an eighteenth-century Austrian example’, American Historical Review 77, 2 (1972), 398–418CrossRefGoogle Scholar; see also ch. 4 of Prolonged connections.

27 Zhao, Z., ‘The demographic transition in Victorian England and changes in English kinship networks’, Continuity and Change 11, 2 (1996), 243–72CrossRefGoogle Scholar. See also a similar, yet less detailed exercise by Anderson using a macro lifetable-based approach. Anderson, M., ‘The social implications of demographic change, 1750–1950’, in Thompson, F. M. L. ed., The Cambridge social history of Britain, vol. II (Cambridge, 1990), 1–71Google Scholar.

28 Zhao, ‘Demographic transition’, 257, 259. Note that Anderson in ‘Social implications’ produces slightly divergent figures for 1851. The difference may be due to the way in which re-marriage is accounted for.

29 Zhao, ‘Demographic transition’, 262–3.

30 Ibid., 256 (Figure 10).

31 Using letters written by paupers dislocated from their parish of settlement due to migration in the first half of the nineteenth century, King has shown how many of the letter writers emphasised a lack of kin support in their requests for financial aid. King, S., ‘Friendship, kinship and belonging in the letters of urban paupers 1800–1840’, Historical Social Research 33, 3 (2008), 249–77Google Scholar. For examples from an earlier period, see Barrett, S., ‘Kinship, poor relief and the welfare process in early modern England’, in King, S. and Tomkins, A. eds., The Poor in England 1700–1850: an economy of makeshifts (Manchester, 2003), 199–227Google Scholar.

32 Office for National Statistics, Households and Household Composition in England and Wales: 2001–11 (2014), available at https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/articles/householdsandhouseholdcompositioninenglandandwales/2014-05-29 [accessed 9 May 2017].

33 P. Laslett, ‘Mean household size in England since the sixteenth century’, in Laslett and Wall, Household and family in past time, 125–58 (436, Table 4.3). This is mistakenly given as 55 per cent by Anderson. Curiously, he also quotes a figure of 55 per cent for individuals living in households of more than six for 1851, so perhaps the two figures became confused. ‘Households, families and individuals’, 424. Note the figures in Table 3 of this article for the period 1650–1821 taken from Wall, are a sub-set of what Laslett termed the 100 ‘English standard’ communities. See note to Table 3.

34 It should be noted that the cohort measure can only be measured every decade, so it cannot be expected to show the same type of curve as the period measures, which can be constructed for every year of age. Also, if the comparable data for 1871 were available, this may change the shape of the cohort ‘curve’. The fact that the cohort line is not in the middle of the period lines for younger ages also suggests that changes for those in younger ages only occurred in the later decades of the period.

35 The current figures on solitary living are estimated from the annual Labour Force Survey. The figures cited in this article are calculated from combining data in Office for National Statistics (ONS), Families and households in the UK: 2016 (London, 2016)Google Scholar, https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2016 [accessed 9 May 2017] and Office for National Statistics, Changes in UK population over the last 50 years (2014), http://webarchive.nationalarchives.gov.uk/20160105160709/http://www.ons.gov.uk/ons/rel/pop-estimate/population-estimates-for-uk--england-and-wales--scotland-and-northern-ireland/2013/sty-population-changes.html [accessed 9 May 2017].

36 Snell, K. D. M.. ‘The rise of living alone and loneliness in history’, Social History 42, 1 (2017), 2–28CrossRefGoogle Scholar, doi: 10.1080/03071022.2017.1256093. Quote from p. 9. This general trajectory in solitary households is also confirmed in Wall, R., ‘Leaving home and living alone: an historical perspective’, Population Studies 43 (1989), 369–89CrossRefGoogle Scholar and in Laslett, P., A fresh map of life: the emergence of the Third Age (London, 1989)Google Scholar. See also Hall, R., Ogden, P. E. and Hill, C., ‘Living alone: evidence from England and Wales and France for the last two decades’, in McRae, S. ed., Changing Britain: families and households in the 1990s (Oxford, 1999)Google Scholar.

37 Snell, ‘The rise of living alone’, Appendix 1; Wall, ‘Leaving home and living alone’, 377.

38 ONS, Families and households 2016, 11.

39 However, these small differences could potentially be an artefact of changes in the way the census office defined the household, as discussed earlier.

40 This point of focusing overly on the household has been made forcibly in particular by Ruggles. See Ruggles, Prolonged connections, Appendix A. See also Ruggles, S., ‘The future of historical family demography’, Annual Review of Sociology 38 (2012), 423–41CrossRefGoogle ScholarPubMed, doi: 10.1145/annurev-soc-071811-145533.

41 Laslett, ‘Mean household size in England since the sixteenth century’, 125–58 (cited at p. 135).

42 These figures are calculated by taking the data from Figures 5a–c (UK population by single year of age, 2014) in ONS, Overview of the UK population: February 2016 (London, 2016)Google Scholar, available at https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/articles/overviewoftheukpopulation/february2016 [accessed 27 May 2017] and comparing them to the data from Figure 6 (People living alone, by age group, 1966–2016, UK) in ONS, Families and households 2016.

43 See, for example, Thomson, D., ‘Workhouse to nursing home: residential care of elderly people in England since 1840’, Ageing and Society 3, 1 (1983), 43–69CrossRefGoogle Scholar. Thompson suggests that pensions led to rapid decline in the numbers of elderly in institutional care. See also the informative discussion on inter-generational dependency and the elderly in Anderson, M., ‘The impact on the familial relationships of the elderly of changes since Victorian times in governmental income-maintenance provision’, in Shanas, E. and Sussman, M. B. eds., Family, bureaucracy, and the elderly (Durham, NC, 1977), 36–59Google Scholar, which draws heavily on the survey findings reported in Booth, C., The aged poor in England and Wales (London, 1894)Google Scholar, and suggests a widespread resistance to parental care in the late nineteenth century.

44 The disqualification for those previously in receipt of poor relief was removed in January 1911 and according to Williams allowed 122,415 individuals (nationally) to move from poor relief to pension during January 1911: Williams, K., From pauperism to poverty (London, 1981)Google Scholar, 171, 207, 212. For an overview of the development of the 1908 Pensions Act, see Collins, D., ‘The introduction of old age pensions in Great Britain’, The Historical Journal 8, 2 (1965), 246–59CrossRefGoogle Scholar; Jones, M., ‘The 1908 Old Age Pensions Act: the poor law in new disguise?’, in Laybourne, K. ed., Social conditions, status and community 1860–c. 1920 (Stroud, 1997), 83–103Google Scholar.

45 However, the absolute numbers of both men and women resident in institutions increased between 1901 and 1911, from 28,245 to 31,796 for women and 34,987 to 36,407 for men, due to improved mortality in older ages.

46 Sellers, E., pensions, ‘Old age and , the “belongingless” poor: a workhouse census’, Contemporary Review 93 (1908), 147–57Google Scholar, cited in Anderson, ‘The impact on the familial relationships of the elderly’, n. 47. See also Crowther, M. A., The workhouse system 1834–1929: the history of an English social institution (London, 1981)Google Scholar, 84 and 219; and Thane, P., Old age in English history: past experiences, present issues (Oxford, 2000), 329Google Scholar for similar views on the dependence of those aged 70 and over on indoor relief due to sickness and medical care.

47 We are indebted to Nicola Blacklaws for information relating to the Poor Law Unions of Stafford and Newcastle-under-Lyme (Staffordshire) and Spalding (Lincolnshire) taken from her ‘The twentieth-century poor law in the Midlands and Wales 1900–c. 1930’ (PhD thesis, University of Leicester, forthcoming).

48 Snell, ‘The rise of living alone’, Figures 3–5, 14–16 (which were generated from the I-CeM data).

49 These typology classifications, based largely on the occupational characteristics of Registration Sub-Districts, have been devised as part of the ‘An Atlas of Victorian Fertility Decline’ project (endnote 3).

50 Reid et al., ‘Gender specific mortality trends over the epidemiological transition’, 73–88.

51 For example, in 2007 David Green, Director of the Institute for the Study of Civil Society was quoted by the BBC as saying, ‘If you take almost any measure – how well children do in school, whether they turn to crime, whether they commit suicide, etc. – it's better to have two parents. It's also the biggest disadvantage of lone parenthood that you're much more likely to be poor.’ See http://news.bbc.co.uk/1/hi/uk/6542031.stm [accessed 27 May 2017].

52 Haskey, J., ‘One-parent families – and the dependent children living in them – in Great Britain’, Population Trends 109 (2002), 46–57Google Scholar. Dependent children are defined as being never-married and aged under 16 or between 16 and 19 (under 19) and undertaking full-time education. These figures contrast to 14 per cent (1961) and 22 per cent (1981) for Great Britain calculated as the percentage of families with a single parent within all households with children in the census of that year. Haskey, J., ‘One-parent families in Great Britain’, Population Trends 45 (1986), 7Google Scholar. However, note also that the figures given in Table 1 of ONS, Families and households 2016 (1996=13.6 per cent lone parent families, 2016=15.4 per cent) vary slightly from the those that have been calculated using the accompanying downloadable dataset and which are given in Table 7 of this article.

53 Wall, ‘Leaving home and living alone’, 374. Wall provides percentages of ‘families of lone parents’ of 22 per cent (1551–1698), 20 per cent (1700–1705), 16 per cent (1752–1796) and 14 per cent (1801–1851), but these are as a percentage of all simple (nuclear) families with children and are therefore not strictly comparable to the figures provided in Table 7. These figures are also produced in Snell, K. D. M. and Millar, J., ‘Lone-parent families and the Welfare State: past and present’, Continuity and Change 2, 3 (1987), 387–422CrossRefGoogle Scholar (392, Table 1 and 414–17, Appendix A).

54 Anderson, M., ‘What is new about the modern family?’, OPCS Occasional Paper 31 (1983), 1–16Google Scholar. See also Anderson, ‘Social implications’, 52–3.

55 An account of Laslett's discovery of the rector's book of Clayworth and its impact has been published in an earlier volume of this journal: Schürer, K., ‘Introduction: Household and family in past time further explored’, Continuity and Change 18, 1 (2003), 9–21CrossRefGoogle Scholar, doi: 10.1017/S0268416003004491. See Gill, H. and Guilford, E. L. eds., The rector's book, Clayworth, Notts. (Nottingham, 1910)Google Scholar and, for the publication of the initial analyses, see Laslett, P. and Harrison, J., ‘Clayworth and Cogenhoe’, in Bell, H. E. and Ollard, R. L. eds., Historical essays presented to David Ogg (London, 1963), 157–68Google Scholar.

56 P. Laslett, ‘Preface’, in Laslett and Wall, Household and family in past time, x. For his defence and retraction of the statement see Laslett, P., ‘The character of familial history, its limitations and the conditions for its proper pursuit’, Journal of Family History 12, 1–3 (1987), 263–84CrossRefGoogle Scholar, esp. 278–9.

57 Ruggles, Prolonged connections, 4.

58 Ibid., 5, Figure 1.1. In n. 10, Ruggles states, ‘Even though the data are scattered, we can be reasonably confident that the peak frequency of extended families occurred sometime after the 1860s.’

59 M. Anderson, ‘Household and the industrial revolution; mid-nineteenth-century Preston in comparative perspective’, in Laslett and Wall, Household and family in past time, 215–35. Laslett, ‘Introduction’, 61, Table 1.3.

60 In an extremely useful article in which Anderson discusses the main findings of his Family structure in the light of subsequent research and newly available data (primarily his National Sample of the 1851 census) he also points to the importance of regional variation in 1851, not only in terms of the overall percentages of households with extended kin, but also the nature of the extension. See Anderson., M. ‘How different were Lancashire families in the Victorian period? Some reflections on another 40 years of research’, in Gritt, A. ed., Family history in Lancashire: issues and approaches (Newcastle, 2009), 43–79Google Scholar.

61 In 1851 1,417 of the 2,176 RSDs were classed as Agricultural (65 per cent), by 1881 this figure was reduced to 61 per cent and to 53 per cent in 1911. In terms of population, the Agricultural RSDs accounted for 45.1 per cent of the national population in 1851, declining to 18.8 per cent by 1911. Likewise the population living in Semi-Rural RSDs declined from 8.4 per cent of the national count in 1851 to 5.6 per cent in 1911. Mining RSDs more than doubled their population share from 3.9 per cent in 1851 to 10 per cent in 1911. Those RSDs designated as Textile declined in their share of the population from 15.9 per cent in 1851 to 9.1 per cent by 1911. The remaining ‘urban’ RSD types (Professional, Semi-Professional, Transport and Other Urban) collectively accounted for 26.6 per cent of the population in 1851, increasing to 56.6 per cent by 1911.

62 Wall, R., ‘The household: demographic and economic change in England, 1650–1970’, in Wall, R., Robin, J. and Laslett, P. eds., Family forms in historic Europe (Cambridge, 1983), 493–512CrossRefGoogle Scholar, here 509. Wall's percentages of extended households are: gentry and clergy (10); yeomen and farmers (18.4); tradesmen and craftsmen (12.1); labourers (agricultural mainly, 10.4) and paupers (11.2).

63 With reference to shopkeepers and artisans, Crossick and Haupt comment that, ‘Most were family enterprises, conceived and run as such however oppressive that may have been for many of its members. Family labour was cheap, readily available, and could be coerced into action with pressures far beyond those to which wage labour was susceptible.’ Crossick, G. and Haupt, H.-G., ‘Shopkeepers, master artisans and the historian: the petite bourgeoisie in comparative focus’, in Crossick and Haupt, Shopkeepers and master artisans in nineteenth-century Europe (London and New York, 1984), 3–31Google Scholar, here 20. Yet in the same volume, writing on Austria, Ehmer comments that ‘only a minority of the master artisans’ households include juvenile or even adult sons’ (p. 199), going on to conclude that ‘the artisans’ traditional mode of life was not in fact as family orientated as is usually assumed’ (p. 212), suggesting that ideals around the artisan family were in part a late nineteenth-century invention. J. Ehmer, ‘The artisan family in nineteenth-century Austria: embourgeoisement of the petite bourgeoisie?’, in Crossick and Haupt, Shopkeepers and master artisans, 195–218.

64 Individual occupations of less than 2,500 households nationally (which account for some 5 per cent of all households) are excluded from this list to avoid a ‘small’ number effect. Other individual occupations in the list include: drapers and linen mercers; clothiers; undefined manufacturers; undefined merchants; nail makers; coal merchants; hatters and warehousemen.

65 It should be noted that these are soldiers in private households, not those recorded as living in barracks, which are classed as institutional in this article.

66 The social class classification used here is that constructed by the Census Office in relation to the 1911 census, the first official social class schema used.

67 Details on occupational specific fertility are available from the ‘An Atlas of Victorian Fertility’ project website (endnote 3).

68 Anderson, ‘Households, families and individuals’, 426; Wall, ‘The household: demographic and economic change in England’, see 499–501, Table 16.4 and discussion.

69 This explains why Anderson numbers grandchildren as the largest co-resident relative group (36.9 per cent).

70 Although the demographic aspects are not discussed, the importance of siblings, especially in middle-class and professional families, is explored in Davidoff, L., Thicker than water: siblings and their relations, 1780–1920 (Oxford, 2012)Google Scholar. See also Davidoff, L., ‘Kinship as a categorical concept: a case study of nineteenth-century siblings’, Journal of Social History 39, 2 (2005), 411–428CrossRefGoogle Scholar.

71 This, of course, is simply in terms of numbers. It is not possible to say anything about the relative amounts the two groups may have contributed.

72 Anderson, M., Family structure in nineteenth-century Lancashire (Cambridge, 1971), 44Google Scholar. See also M. Anderson, ‘Household structure and the industrial revolution: mid-nineteenth-century Preston in comparative perspective’, in Laslett and Wall, Household and family in past time, 215–35. But note the proportion of households with co-resident relatives is not the same as the proportion of extended households. It will always be slightly higher since not all households with relatives have a core family (CFU).

73 See also Anderson, ‘How different were Lancashire families’ in which he stresses how Preston and other Lancashire cotton towns, as well as textile towns in the West Riding of Yorkshire stand out as atypical in terms of their family structures in 1851.

74 In Preston in 1851 Anderson notes that ‘in 14% of all cases where the mother worked (17% of all cases where she worked in a factory) the house contained an otherwise unemployed grandmother. Most of these would have been available as guardians.’ Anderson, Family structure, p. 74. This was in contrast to the potteries where the majority of female workers were younger, unmarried women. See Dupree, M. W., Family structure in the Staffordshire potteries 1840–1880 (Oxford, 1995), 202–03CrossRefGoogle Scholar.

75 Garrett, E. M., ‘The trials of labour: motherhood versus employment in a nineteenth-century textile centre’, Continuity and Change 5, 1 (1990), 121–54CrossRefGoogle Scholar.

76 In the case of extended households, for those born in England and Wales for whom distance between place of birth and place of enumeration can be calculated, the I-CeM database reveals that the proportion of those born ten or more miles from where they were living was 26.4, 30.6, 35.4, 34.5, 34.6 and 35.9 in each of the census years 1851–1861 and 1881–1911 for extended kin. In comparison, for members of the core family group the percentages born ten miles or further were 22.6, 26, 29.4, 28.4, 28.4 and 30.5. Even when filtering out those aged under 12 (those for whom the process of leaving home was largely yet to start) the difference between core family members and those forming the extension was marginal, on average just 3 km. The mean distances (km) between place of enumeration and birth for core family members age 12 or over by census year were 15.5, 19.1, 22.4, 22.3, 22.5 and 23.7. For extended kin aged 12 or over the comparable figures were 18.1, 22.4, 25.9, 25.9, 25.9 and 26.8.

77 The discussion of inter-generational dependency in Anderson, ‘The impact on the familial relationships of the elderly’ is also useful in relation to this suggestion, arguing that for successful co-residence mutual benefit is essential.

78 This point is demonstrated by Anderson, ‘Social implications’, 49, Table 1.5.

79 Investigating the Fathercraft and Fathers’ Councils movements for the inter-war period of the twentieth century, which campaigned for fathers to have a more active role in child support and caring, Fisher suggests that the general perception in this period was that fathers were largely incapable of looking after infants and young children alone. Fisher, T., ‘Fatherhood and the British Fathercraft Movement, 1919–39’, Gender & History 17, 2 (2005), 441–62CrossRefGoogle Scholar, here 452.

80 It should be noted that re-marriage is not easy to identify in censuses of England and Wales, 1851–1911. No information on parity of marriage is given. If a widow re-married, any resident children of hers may be identified as stepchildren. If a widower re-married then this is almost impossible to detect.

81 It is, however, impossible to tell from the census if children resident in boarding schools had parents alive or not.

82 Pooley and Turnbull have suggested that age of leaving home fell during the nineteenth and early twentieth centuries. The individual level census data demonstrate that this could not have been the case (at least for the period 1851–1911) and their result is most probably an artefact of using life history data that fail to capture all residential moves. See Pooley, C. and Turnbull, J., ‘Leaving home: the experience of migration from the parental home in Britain since c. 1770’, Journal of Family History 22, 4 (1997), 390–424CrossRefGoogle Scholar. For the pre- and early nineteenth century when age at leaving home seems earlier than the figures provided here – mainly due to higher levels of farm service and apprenticeships – see Snell, K. D. M., Annals of the labouring poor: social change and agrarian England, 1660–1900 (Cambridge, 1985), 322–32CrossRefGoogle Scholar; Wall, ‘Leaving home and the process of household formation’ and Wall, R., ‘The age at leaving home’, Journal of Family History 3, 2 (1978), 181–202CrossRefGoogle Scholar.

83 The calculation of the singulate mean age at leaving home is based on Hajnal's classic formulation of the Singulate Mean Age at Marriage (SMAM). For a description and discussion, see Schürer, K., ‘Leaving home in England and Wales, 1850–1920’, in van Poppel, F., Oris, M. and Lee, J. eds., The road to independence: leaving home in Eastern and Western societies, 16^th–20^th centuries (Berne-Bruxelles, 2003), 33–84Google Scholar. This chapter also illustrates that there were most probably important geographic variations around the general national trend indicated here, both in terms of timing and extent.

84 The work of Snell and Wall for an earlier period suggest that if one could construct a curve showing the proportion of children not living with a parent by age for the pre-1851 period it would be to the left of that shown for 1851 in Figure 7 due to the prevalence of farm service and apprenticeships in particular – however, differing parental mortality also needs to be taken into consideration. See Snell, Annals of the labouring poor (esp. Table 7.2) and Wall, ‘Leaving home and the process of household formation’ comparing Table 2 (p. 88) showing headship rates for the four communities of Ealing (1599), Ardleigh (1796), Winwick with Hulme (1801) and Chilvers Coton (1684 and 1781), with Figure 1 (pp. 84–5) showing the proportion of children with their parent(s) for the same places.

85 But note for a selection of mainly midland and south-western agricultural parishes, using settlement examinations, Snell estimates that in the eighteenth century, departure from home for boys was slightly earlier than girls for the period 1700–1815 but the pattern was reversed thereafter (Snell, Annals of the labouring poor, 326). This is in contrast to Wall who in comparing the sex ratios of those older than ten for 24 communities (with three places repeated) for the period 1599–1831, states that ‘it was rather rare in the English experience that sons would remain in the parental home in preference to daughters’: Wall, ‘Leaving home and the process of household formation’, 94.

86 Illegitimate fertility rates calculated as part of the ‘An Atlas of Victorian Fertility Decline’ project (see endnote 3) suggest a decline from 18.45 births per 1,000 non-married women (aged 15–49) in 1851 to 7.45 per 1,000 in 1911, in particular falling sharply between 1861 (18.07/000) and 1901 (8.10/000). Overall the ratio of illegitimate births of all births fell from 0.068 in 1851 to 0.041 by 1911. However, a key feature of this decline is not just the pace of the decline, but the extent to which rates by 1911 became highly uniform across England and Wales in comparison with 1851 when there was considerable geographic variation in illegitimate fertility rates. See also, Teitelbaum, British fertility decline, 151, Table 6.10a. However, for the parish register period (c. 1550–1837) it has been observed that illegitimate fertility mirrors legitimate fertility and varies inversely with age at marriage (and vice versa), placing an emphasis on courtship intensity rather than time spent unmarried. It may be that with rising age at leaving home, this relationship was broken in the latter decades of the nineteenth and early twentieth centuries. See, for example, Adair, R., Courtship, illegitimacy, and marriage in early modern England (Manchester, 1996)Google Scholar; Laslett, P., ‘Long-term trends in bastardy in England’, in Laslett, P., Family life and illicit love in earlier generations (Cambridge, 1977), 89–115CrossRefGoogle Scholar; Laslett, P., ‘Introduction: comparing illegitimacy over time and between cultures’, in Laslett, P., Oosterveen, K. and Smith, R. M. eds., Bastardy and its comparative history: studies in the history of illegitimacy and marital nonconformism in Britain, France, Germany, Sweden, North America, Jamaica and Japan (London, 1980), 1–65Google Scholar.

87 Forster's Elementary Education Act of 1870 provided for the public education of those aged 5 to (under) 13. More significant, however, was the 1880 Elementary Education Act, which required local authorities to pass by-laws making attendance compulsory, in effect by introducing penalties for those aged under 14 being illegally employed. Clearly, however, this was not uniformly enforced (or enforceable). Also important was the Act of 1891, which provided state funding to School Boards, making elementary education free. The 1902 Education Act, while important in a number of respects, including replacing School Boards with Local Education Authorities, did not raise the (compulsory) school leaving age. This was done in 1918 under Fisher's Education Act. However, the situation regarding the employment of children was complicated by the fact that this was governed both through a number of occupation-specific regulations and a host of local by-laws. See Keeling, F., Child labour in the United Kingdom: a study of the development and administration of the law relating to the employment of children (London, 1914)Google Scholar, esp. xi–xxxii, 56–9, in which he states that: ‘Whatever may have been the intentions either of the draughtsmen of the [Education] Acts or of the Parliaments which passed them, it is certain that in fact the Acts have not availed to prevent the employment of children attending school out of school hours. It has not even been possible to enforce the apparently unconditional minimum age of 10, which was established by the English Act of 1876 and the Scottish Act of 1878’ (p. xxi). The I-CeM data record the following percentages with a ‘working’ occupation for those aged 12 to 14, 1851–1861 and 1881–1911: boys 51, 50, 38, 42, 37 and 32; girls 32, 31, 24, 25, 21 and 19 per cent.

88 For a discussion of regional variations in household structure in the pre-1851 period, see Schürer, K., ‘Variations in household structure in the late seventeenth century: toward a regional analysis’, in Schürer, K. and Arkell, T. eds., Surveying the people: interpretation and use of document sources for the study of population in the later seventeenth century (Oxford, 1992), 253–78Google Scholar; Wall, R., ‘Regional and temporal variations in English household structure from 1650’, in Hobcraft, J. and Rees, P. eds., Regional aspects of British population growth (London, 1979), 89–113Google Scholar. Analysis for the post-1851 period also indicates significant geographical variation: Schürer, K. and Penkova, T., ‘Creating a typology of parishes in England and Wales: mining 1881 census data’, Historical Life Course Studies 2 (2015), 38–57Google Scholar, doi: http://hdl.handle.net/10622/23526343-2015-0004?locatt=view:master; Wall, R., ‘Regional and temporal variations in the structure of the British household since 1851’, in Barker, T. and Drake, M. eds., Population and society in Britain, 1850–1980 (London, 1982), 62–99Google Scholar.