85 YEARS OF US RURAL WOMENS TIME USE

85 YEARS OF US RURAL WOMENS TIME USE

85 YEARS OF US RURAL WOMENS TIME USE Teresa Harms and Jonathan Gershuny Centre for Time Use Research University of Oxford ACKNOWLEDGEMENTS: We gratefully acknowledge the financial support of the UK Economic and Social Research Council (grant references ES/L011662/1; ES-060-25-0037, ES-000-23-TO704 and ES-000-23-TO704A), the European Research Council (grant reference 339703) and the University of Oxford John Fell Fund. Outline of presentation The presentation falls into two parts: 1. An outline of the procedures used to revive the micro-data from the 1920s USDA diary studies of womens time use. 2. A first application of this material, comparing it with the rural women from 1975 Michigan National time diary study and the ATUS 2003-10, to reconsider the Vanek/Cowan housework still takes time thesis. 2 Background to study The Morrill Land Grant College Act of 1862 supported the establishment of US state colleges for agriculture and the Hatch Act (1887) the development of state agricultural experiment stations. The Smith-Lever Act (1914) created a Cooperative Extension Service associated with each US land-grant institution. The 1925 Purnell Act, with a focus on agricultural economics and rural sociology, enabled the US Department of Agriculture (USDA) to collect evidence of farm womens time use across a number of state agricultural experiment stations.

Dr Hildegard Kneeland, a Home Economist employed by the US Bureau of Economics conducted diary surveys from farm, town and college women for comparative purposes. Her 1929 Annals article sets out a modern case for accounting extensions to recognise womens unpaid domestic work. Published tabular results from the original USDA researcher-produced summaries has been used (in Vanek 1974, 1978; Ramey 2009) but the individual-level micro-data have not been available until now for analysis. 3 Materials so far recovered Originally >1500 women farm and town were surveyed across the United States. The recovered material includes: Researcher-produced diary-based weekly minutes summary tables in 60 activities, plus weekly help totals in various specific household tasks) for 566 farm and other rural women mainly from New York, California and Michigan States. Seven-day own-words diaries, detailed supplementary questionnaires (household composition and characteristics, appliances, equipment) and researcher-produced summary sheets for 77 College women (Kneelands study). We have digitised all of the recovered records by scanning them and entering the data into a large database ready for analysis (SPSS, STATA). 4

Examples of 7-day own-words diary (1930) Example of USDA researcher-produced summary record 6 Methods for identifying womens individual and household characteristics Family and given names/initials and address were attached to the 566 USDA researcher-prepared aggregate summaries. These were matched to records in Ancestry.com including: US Federal Census micro-data to 1940 (70-year embargo in US) Birth, Death & Marriage Indexes Voting Registers (mainly Californian) Social Security Numbers City Directories Military records (including drafts) Immigration & travel documents (passport applications, etc.) Other material (obituaries, newspaper articles, photographs, etc.) 7 US Federal Census 1930 Census database variables

For diarist and husband data from two censuses (mainly 1920 & 1930). Farm & town womens family characteristics derived from household census record (husband head of household in 1920 census). 94% of aggregate records matched to census records. Census Roll and Enumeration District Children living in household (name/age/sex) Address at Censuses and USDA Survey Other household members (parents/step-children/othe r family/ servants/lodgers) Rent/own/live in parents or relatives home/other Place of birth (diarist, husband and parents)

Live on farm? Own radio? (1930 census) Race/immigrant/dates of immigration/naturalised?/ mother tongue War veteran (WW1/Spanish American War) Occupation & industry Birth & Death Index Employed at Census? Maiden name of diarist Attended College at Census? Remarks Farm or town samples? Is the diarist's husband a farmer? Not (farmer in neither census) Possibly (farmer in

one census) domicile, nearest census: not recorded not living on farm living on farm total Farmer by occupation? Probably (farmer both censuses) Total 110 32 1 53 144 10 178 1 173 354 142

198 188 528 27% 38% 36% 33% 67% N and age-distributions, US diary samples Age Group 16-19 20-29 30-39 40-49 50-59 60-65 Total N N of days USDA N 1920s 3 70 203 162

67 22 1 528 3696 AHTUS rural women %________________________________ 1920s 1975 20031965 1985 0.6 2.9 0.7 2.0 0.8 13.3 27.4 16.7 26.9 19.9 38.5 21.9 25.8 24.9 27.2 30.7 16.8 26.0 20.9 22.2 12.7

20.8 22.2 21.7 23.4 4.2 10.3 8.7 3.6 6.5 Missing 249 982 261 5331 The housework time paradox The claim that dissemination of domestic labour saving equipment does not in fact save labour is very durable. (Mokyr (2000) calls this finding the Cowan paradox, after Schwartz-Cowan (1980)tho her book does not make this claim.) Source is a 1974 Scientific American article based on tables from the USDA diaries. Opening sentence: As one might expect, working women spend less time in housework than their mothers and grandmothers did some 50 years ago. Women who are not in the labor force however, spend just as much time. Vanek (1974, p. 116, our emphasis) However once she turns to discussing the evidence, her term shifts to household workwhich includes childcare and shopping! Irrespective of terminology, we see two key problems with her finding: A potential selection bias effect that results from Vaneks focus on

women who are not in the workforce. The X-section difference = change fallacy (not explored here). A selection bias effect A process leading to historical growth in household workload of both employed and non-employed women Assume, other things being equal: womens Pempl inversely related to unpaid workload growing acceptability of womens paid work i.e. Pempl increases over time at each workload level The mean workload level of non-employed women rises over historical time (as does that of employed women). Both groups have on average more unpaid work to do even if total workload remains unchanged (because of proportional shift into employment). What do we expect to find? Vaneks claims (1920s to 1965): Increase in household work time for non-employed women. A difference in housework trends between employed and non-employed women. We are looking for: Longer-term evidence that labour-saving domestic equipment fails to save unpaid labour time. 300 Figure 1a US farm and small town married women 1920s to 2000s:

Decreasing work-time categories . 95% confidence intervals 250 Figure 1b US farm and small town married women 1920s to 2000s: Increasing work-time categories 95% confidence intervals 250 200 cooking, farm or paid work clearing, cleaning 200 150 minutes per day minutes per day 150

100 100 shopping 50 50 care laundry sewing accounts utilities 0 1920s child and adult 1975 0 2000s 1920s 1975 2000s 300

350 Figure 2a US farm and small town married women 1920s to 2000s, seven or fewer hours paid or farm work per week. Figure 2b US farm and small town married women 1920s to 2000s, more than seven hours paid or farm work per week. 300 250 farm or paid work 250 200 cooking, clearing, cleaning minutes per day 200 minutes per day 150 cooking, clearing, cleaning 150 adult and child care

100 100 shopping etc. mending, knitting, sewing shopping etc. farm or paid work 50 50 laundry 0 1920s accounts utilitie s 1975 2000s child and adult care mending,

knitting, sewing 0 1920s laundry accounts utilities 1975 2000s Rural married womens unpaid work minutes per day Multiple R (age & squared, N kids, age youngest child, interactions) economically active econ. active*1975 econ. active*2003Surveyed 1975 Surveyed 2003-11 (Constant) OLS models p= .0005 *** p=.005 ** p=.05 * other clothes shop, cooking domestic care admin 0.46 0.31

0.34 0.16 .. -2.3 -29.0 -14.0 -40.7 -77.3 96.2 *** * *** *** *** -9.1 -33.4 -27.7 -44.2 -74.7 126.3 ** ** *** *** *** . -28.4

7.0 21.6 -65.1 -83.5 87.5 *** *** *** *** *** . -3.8 -15.8 -8.0 69.8 76.8 25.7 *** *** care of persons 0.50 all unpaid 0.42 .

-10.5 -17.2 -24.3 24.6 36.5 -18.7 . -54.1 -88.6 -52.5 -55.6 -122.2 317.1 * * * ** ** ** ** *** *** 300 250 Estimated Cooking and Cleaning Time US married rural women aged 40

US married rural women aged 40 200 250 3 children youngest 3 3 children youngest 15 1 child aged 3 1 child aged 15 no children 3 children youngest 3 3 children youngest 15 1 child aged 3 1 child aged 15 3, 3 150 minutes per day 200 minutes per day Estimated Child and Adult Care Time 1, 3 1, 3

3, 15 3, 15 100 150 3, 15 3, 15 none 50 100 1, 3 none 50 3, 3 0 1, 15 1, 15 550 Figure 4 Estimated Total Unpaid Work Time US married rural women aged 40

3, 3 500 1, 3 450 3, 15 minutes per day 400 3, 3 1, 15 1, 3 350 3, 15 300 1, 15 250 none 200 3 children youngest 3

3 children youngest 15 1 child aged 3 1 child aged 15 no children none Why are these changes happening? Is it because of change in population characteristics (e.g. age distributions, age and N of children, proportion of women in employment)? Is it change in behavioural propensities (i.e. people with particular characteristics do different things with their time.) Or is it both changing in the same direction at the same time? Oaxaca decomposition of OLS regression historical change in time in an activity = intercept change effects + coefficient change effects + means change effects + interaction change effects Dependent Yit=1-Yit=0 , n independent vars in an OLS regression: intercept effects= intit=1 - intit=0 coefficient effects= sumj=1n(ijt=0 * (bijt=1 - bijt=0)) mean effects= sumj=1n(bijt=0 * (ijt= - ijt=0)) interaction effects= sumj=1n((bijt=1 - bijt=0)*(ijt=1 - ijt=0)) behavioural propensity change = intercept effects + coefficient effects Bianchi et al (2001 p.211) Proportion of all historical change in time use related to behavioural propensities

cooking other domestic clothes care shopping child & adult care 1920s-1975 1975-2000s 90% 94% 120% 83% 96% 97% 110% 129% 139% 111% 1920s-2000s 95% 173% 99% 105% 129% Findings In relation to the Vanek thesis: Decreasing total household work time for almost all groups of women. No difference in trends between employed and non-employed. Dominance of behavioural

change consistent with common-sense expectation from labour-saving equipment. Findings In relation to the Vanek thesis: Decreasing total household work time for almost all groups of women. No difference in trends between employed and non-employed. Dominance of behavioural change consistent with common-sense expectation from labour-saving equipment. In relation to 1920s USDA timediary data: The addition of census and other contextual data to microlevel diary material produces a useable dataset. .which (partially) extends the AHTUS backwards by 40 years. References

Bianchi, Suzanne M., Melissa A. Milkie, Liana C. Sayer & John P. Robinson (2000) Is anyone doing the housework? Trends in the gender division of household labor, Social Forces 79 (1): 191-228. Kneeland, Hildegarde (1929) Woman's economic contribution in the home, Annals of the American Academy of Political and Social Science, 143: 33-40. Mokyr, Joel (2000) "Why "More Work for Mother?" Knowledge and Household Behavior, 1870-1945." The Journal of Economic History 60 (1): 1-41. Cowan, Ruth Schwartz (1983) More work for mother: The ironies of household technology from the open hearth to the microwave. New York: Basic Books. Ramey, Valerie. A. (2009) Time spent in home production in the Twentieth-Century United States: New estimates from old data, The Journal of Economic History, 69 (1): 1-47. Vanek, J. (1978) Household technology and social status: Rising living standards and status and residence differences in housework, Technology and Culture, 19 (3): 361-375. Vanek, J. (1974) Time spent in housework, Scientific American, 5 (231):116-120.

Recently Viewed Presentations

  • Screening for Bladder Cancer: What You Need to Know

    Screening for Bladder Cancer: What You Need to Know

    Of the 3 methods used, white-light cystoscopy is the preferred method. But while it is considered the "gold standard" for diagnosing bladder cancer, the traditional method of white-light cystoscopy has limitations. White-light technology does help doctors see abnormalities; but white...
  • Prof. Dr. Recep BOZLAAN 2 LDERLK NE DELDR

    Prof. Dr. Recep BOZLAAN 2 LDERLK NE DELDR

    Prof. Dr. Recep BOZLA─×AN
  • What Are Natural Disasters?

    What Are Natural Disasters?

    The world is always changing. Natural disasters are changes which are so great they may cause damage to the shape of the land or the lives of people and other things. Great changes happen deep inside the Earth and on...
  • In-Home Services DCBS Referral and Selection Contacts

    In-Home Services DCBS Referral and Selection Contacts

    In-Home Services: DCBS Referral & Selection Contacts Central Office Contacts Intensive Family Preservation Services (IFPS), Reunification Services (FRS), Families and Children Together Safely (FACTS): Lynda Robertson & Charity Roberts (502) 564-2136
  • ACSI Teacher Certificate Q&A Session/Information Fair

    ACSI Teacher Certificate Q&A Session/Information Fair

    with Association of Christian Schools International (ACSI) since August 4, 2014. Graduates will receive the Professional Teacher Certificate of ACSI, recognized by ACSI member schools (nearly 24,000 schools as of 2014) and Christian schools all over the world.
  • The Models of Communication - amizone.net

    The Models of Communication - amizone.net

    Linear Model. In this linear communication model, communication is like giving an injection: a sender encodes ideas and feelings into some sort of message and then conveys them by means of a channel (speech, writing, and so on) into a...
  • Big Data: Technologies, Opportunities, and Challenges

    Big Data: Technologies, Opportunities, and Challenges

    CMPS 490 / CSCE 598Big Data Analytics. Description: Essentials of Big Data analytics. Topics include: challenges and opportunities posed by Big Data in a variety of domains, predictive analytics or other advanced methods to extract value from data, innovative statistical...
  • Staffing (Recruitment, Placement and Compensation) SSG Matthew Shifflett,

    Staffing (Recruitment, Placement and Compensation) SSG Matthew Shifflett,

    EDP and HDP are special pay options for technicians working in hazardous duties. Same concept, just different acronym for each pay system. Technician's PD determines if eligible, and OPM determines rate of compensation (i.e. 10% of base rate if work...