Insights from time use research and mixed data methods. A - - PowerPoint PPT Presentation

▶
insights from time use research and
SMART_READER_LITE
LIVE PREVIEW

Insights from time use research and mixed data methods. A - - PowerPoint PPT Presentation

Apr 09, 2023 26 likes •250 views

Insights from time use research and mixed data methods. A contribution to INSIGHTS: bringing together sensor technology and social research J Gershuny, London, 20 July 2015 Research funded by the UK Economic and Social Research Council, the

slide-1
SLIDE 1

Insights from time use research and mixed data methods.

A contribution to INSIGHTS: bringing together sensor technology and social research

J Gershuny, London, 20 July 2015 Research funded by the UK Economic and Social Research Council, the European Research Council, and the US National Institutes of Health National Cancer Institute

slide-2
SLIDE 2

In what follows…

  • How to measure activities?
  • Time diary research.
  • Mixed diary and sensor research:
  • 1. …to simplify diary data collection
  • 2. …to validate and calibrate diaries
  • 3. …to amplify the meaning of diary evidence
  • An example: daily metabolic activity (METs)
  • Situating sensors: sampling, representativeness
slide-3
SLIDE 3

Measuring time use

  • ‘Beeper’ sampling (ESM/EMA) studies:

– What were you doing when the beeper went?

  • Questionnaire ‘stylized’ items:

– “How much time did you spend … last week?” – “How often do you….? How often …last month?”

– respondents don’t know….and will exaggerate!

  • Diaries:

– What were you doing at 4am? Doing anything at the same time? Who were you with? What did you do next? What time did that start? …..

slide-4
SLIDE 4

UK HETUS 2014/5 diary

slide-5
SLIDE 5

Why diaries are the gold-standard for activity research

Diary methods are essentially preferable to questionnaire:

  • “yesterday” format maximises recall accuracy & specificity
  • sequential nature of the diary record provides

– prompts (this activity followed that) and – cues for recall effort (from gaps in the continuity of reconstructions)

  • the neutral nature of the request for diary record gives no

idea of researchers’ interests (versus directive effect of the particular activities identified by PA questionnaire items)

  • the real-time 24-hour constraint of the diary, means that

misrepresentation is more difficult than truthsaying. It is simpler for respondents to provide truthful diary accounts!

slide-6
SLIDE 6

Random-day diaries vs “stylised” q’nnaire items: participation estimates

Gershuny (Ann. Ec. & Stat. 2012)

slide-7
SLIDE 7

Brief history of time diary research History – Russian peasant studies c. 1900 – M. Pember-Reeves, UK, Fabian Soc.,1911 – I. Strumilin, Russia, NEP 1921 – H. Kneeland, USA, USDA 1925 – A. Szalai, UNESCO 1965 – Harmonised European Time Use Study (HETUS) 1999-2015 – American Time Use Study (ATUS) 2003—

slide-8
SLIDE 8

Major time diary surveys in MTUS

(parentheses=studies under negotiation or preparation)

1960s 1970s 1980s 1990s 2000s 2010s

Australia

1974 1987 1993, 1997 2006

Canada

1971-2 1981, 1986 1992, 1998

China

2000

Denmark

1964 1987 2001 (2011)

Finland

1979 1987-8 1999-2000 (2011)

France

1966 1974-5 1988 1999 2009 2010

Germany

1965-6 1990 2001-2 (2012)

Hungary

1965 1976-7

Italy

1979 1989 2002-3 (2010-11)

Netherlands

1975 1980, 1985 1990, 1995 2005 (2011)

Norway

1971-2 1981-2 1990-1 2000-1 (2010-11)

Slovenia

1965 2001-2

South Africa

2000

South Korea

1997 2003,2008 (2014)

Spain

1992, 1997 2002-3 2010

Sweden

1990-1 2000-1 (2010-11)

United Kingdom

1961 1974-5 1983, 1987 1995 2001, 2005 (2014-15)

United States

1965-6 1975-6 1985 1992-4, 1998 2003-9 2010-12

slide-9
SLIDE 9

CTUR diary/instrumentation

  • 6 towns GPS/GSM, 2012-13

– With Trajectory Partnership – 1300 days multiple activity, copresence, enjoyment

  • CAPTURE24, 2015-16

– With Oxford U. Public Health Dept – Diary, camera, accelerometer. 150 days’ data

  • Energy-24, 2016-

– With Oxford Sustainable Energy Institute – Diary, camera, accelerometer + energy meter.

slide-10
SLIDE 10

CAPTURE24, energy-24

slide-11
SLIDE 11

NATURE Features Editor’s day

Helen Pearson “The Time Lab” NATURE, Vol 526, 22 October 2015, pp 492-496

slide-12
SLIDE 12

Results from CAPTURE24 Pilot Study

Developing a Method to Test the Validity of 24 Hour Time Use Diaries Using Wearable Cameras: A Feasibility Pilot, Paul Kelly, Emma Thomas, Aiden Doherty, Teresa Harms, Órlaith Burke, Jonathan Gershuny, Charlie Foster, PLOS ONE | DOI:10.1371/journal.pone.0142198 December 3, 2015

slide-13
SLIDE 13

METs & time-use diaries

  • Ainsworth Compendium Tudor Locke 2009

– Attaching METs to ATUS – Enables balance between “exercise” & other activities

  • Van der Ploeg et al 2010

– “TUS used retrospectively …physical & sedentary activity”

  • Ng and Popkin 2012

– Historical change in physical activity – Attaching METs to MTUS data (UK and US) – …. METs to non-diary data (China, India, Brazil) – Uncover dramatic growth in sedentary activity

slide-14
SLIDE 14

Two diary-based physical activity measures

  • 1. Aggregate daily metabolic activity (METs)

Mean daily MET per minute score (mdMETs), respondent with N activities during sampled day

mdMETs = ( (MET activity1 * Dur activity1) +(MET activity2 * Dur activity2) +....(MET activityN * Dur activityN) )/1440

  • 2. Metabolic activity by duration.

For public health recommendations of the form: “…>150 minutes a week moderate intensity PA…..” reorder each diarist’s daily activities from the highest to lowest intensity level, then calculate threshhold proportions.

slide-15
SLIDE 15

ATUS/MTUS METs comparison 20% reduction in total variance

1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 15-30 30-45 45-60 60-75 75

primary incomplete high high school grad incomplete college Batchelors degree Masters degree Doctorate, professional Fig 2b: Mean daily METS: ATUS 6-digit categories Men, by age and educational attainment Age group

slide-16
SLIDE 16

METs and self-reported health

Simple OLS, entire good-diarist samples

UK HETUS 2001 ATUS 2006-12

20,107 days 86,196 days

R Squared

0.135 0.084 mean daily METs

2.156 *** 1.124 ***

....squared /1000

  • 0.531

***

  • 0.309

***

age

0.123 *** 0.093 ***

age *daily METs

  • 0.110

***

  • 0.086

***

.........squared

0.026 *** 0.021 ***

(Constant)

  • 0.725

1.102 ***

slide-17
SLIDE 17

UK/US METs and Health

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 1.1 1.3 1.5 1.7 1.9 2.1 2.3 One day model UK aged 30 One day model UK aged 70 Mean daily METs negative-scored subjective health 1.0 1.5 2.0 2.5 3.0 3.5 4.0 1.1 1.3 1.5 1.7 1.9 2.1 2.3 One day model US aged 30 One day model US aged 70 Mean daily METs negative-scored subjective health

slide-18
SLIDE 18

Exercise: a balanced perspective

slide-19
SLIDE 19

Contributions to daily METs, ordered by activity level

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137

Poland sample ages 15+

10-minute intervals ordered by METs

2 10 18 26 34 42 50 58 66 74 82 90 98 106 114 122 130 138

UK sample ages 15+

exercise travel paid work unpaid work child and adult care leisure away from home leisure at home sleep, personal care 10-minute intervals ordered by METs

1 16 31 46 61 76 91 106 121 136 151 166 181 196 211 226 241 256 271 286

US sample ages 15+

5-minute intervals ordered by METs

slide-20
SLIDE 20

METograms: % of sample with aggregate METs levels for various durations on a diary day

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0 20 40 0 20 40 0 20 40 0 20 40 0 20 40 0 20 40 0 hrs 1 hr 2 hrs 3 hrs 4 hrs 5 hrs

Polish men aged 18-39

0 20 40 0 20 40 0 20 40 0 20 40 0 20 40 0 20 40 0 hrs 1 hr 2 hrs 3 hrs 4 hrs 5 hrs

UK men aged 18-39

>=6 METs 5-6 METs 4-5 METs 3-4 METs 2-3 METs 1-2 METs

0 20 40 0 20 40 0 20 40 0 20 40 0 20 40 0 20 40 0 hrs 1 hr 2 hrs 3 hrs 4 hrs 5 hrs

US men aged 18-39

slide-21
SLIDE 21

Research Opportunities

  • Combine diaries with “physical methods”

for validation, calibration and adjustment

  • Recruit study participants from large

random sample “feeder” surveys, allowing “inverse participation probability weights” producing nationally representative stats.

  • Use representative—and comprehensive—

time use data to provide a framework for interpreting the accurate but partial views

  • f daily life derived from passive sensors.
slide-22
SLIDE 22
  • B E Ainsworth., W. L. Haskell, S. D. Herrmann, N. Meckes, D. R. Bassetr JR., C. Tudor-Locke, J.
  • L. Greer, J. Vezina, M. C. Whitt-Glover, and A. S. Leon, ‘Compendium of Physical Activities: A

Second Update of Codes and MET Values’. Med. Sci. Sports Exerc., Vol. 43, No. 8, 2011, pp. 1575–1581.

  • Jonathan Gershuny, ‘Too Many Zeros: a Method for Estimating Long-term Time-use from

Short Diaries.’ Annals of Economics and Statistics. 105-106 January-June 2012 pp 247-271.

  • Victor Kipnis, Douglas Midthune, Dennis W. Buckman, Kevin W. Dodd, Patricia M. Guenther,

Susan M. Krebs-Smith, Amy F. Subar, Janet A. Tooze, Raymond J. Carroll, and Laurence S. Freedman ‘Modeling Data with Excess Zeros and Measurement Error: Application to Evaluating Relationships between Episodically Consumed Foods and Health Outcomes’ , Biometrics 65, December 2009 pp 1003–1010 .

  • S.W. Ng and B, M. Popkin ‘Time use and physical activity: a shift away from movement across

the globe’ Obesity Reviews 13, (2012) 659–680.

  • Catrine Tudor-Locke,Tracy L.Washington, Barbara E. Ainsworth, and Richard P. Troiano

‘Linking the American Time Use Survey (ATUS) and the Compendium of Physical Activities: Methods and Rationale’ Journal of Physical Activity and Health (2009).

  • Hidde P. van der Ploeg, Dafna Merom, Josephine Y. Chau, Michael Bittman, Stewart G. Trost,

and Adrian E. Bauman ‘Advances in Population Surveillance for Physical Activity and Sedentary Behavior: Reliability and Validity of Time Use Surveys’ American Journal of Epidemiology (2010) pp 1999—1206

References