Innovation in online data collection for scientific research The Dutch MESS project Marcel Das

1
Innovation in online data collection for
scientific research The Dutch MESS
projectMarcel Das
2
MESS Project

• Purpose to build an advanced data collection
  environment for the social sciences
• Maximal opportunities for innovation
• Multidisciplinary
• Fast
• Cost effective
• Easy to use for everyone in the scientific
  community

3
Central element MESS project the LISS panel

• Online panel of 5,000 households
• 8,000 individuals (gt 16 years)
• Questionnaires each month, 30 min.
• Incentive 15 ( ? 21,500) per hour
• At no cost if for scientific use

4
LISS panel

• Online interviews as method, but
• Probability sample drawn from address sampling
  frame of Statistics Netherlands
• Includes households without Internet access (less
  than 15) CentERdata provides equipment
• Contacted by letter, telephone or visit

5
simPC

• Very small and silent
• Only the most frequently used functions
• Automatic maintenance, virus protected
• Simple operation and readable screens
• Installation and support

6
Non-response patterns

• Similar to those of other leading scientific
  panels
• Superior to commercial access and volunteer
  panels- no coverage problems - no
  self-selection

7

• Attention for attrition
• Refreshment sample in 2009 (stratified) and 2011
  (random), in close collaboration with Statistics
  Netherlands

8
Number of registered households by subsamples
Jan 2008 - July 2012
Source Scherpenzeel et al. 2012
9
Non-Internet households(Leenheer and
Scherpenzeel, 2012)

• Who are they?
• Do we get them in the panel?
• Are participants representative?
• Do they stay in the panel?
• Do they stay representative or do they start to
  use the Internet more frequently?

10
Results of study

• Differ in many ways from Internet hh
• Difficult to get into the panel
• In panel representative with respect to age,
  household size and origin
• Loyal and reliable panel members
• About half remain Non-Internet
• Using Internet does not change the answers to
  questions

Who are they? Get them in the
panel? Represen-tative? Stay in the
panel? Stay repre-sentative?
11

• Use of the LISS panel
• New data collection
• Academic researchers, irrespective of
  nationality, can use the LISS panel at no cost
• Proposals can be submitted throughout the year
  (see http//www.lissdata.nl)
• Available data (free of charge)
• Proposed studies
• Longitudinal core study

12
Longitudinal Core Study

• Questions were designed in close collaboration
  with experts in the relevant fields
• Core study borrows from various national and
  international surveys (to facilitate comparisons
  with other data sources)
• Topics Household and family, Economic situation
  and housing, Work and schooling,
• Social integration and leisure, Health,
  Personality, Religion and ethnicity, Politics and
  values
• Wave 5 Nov 2011 to June 2012

13
Innovation in data collection methods
14
Advanced weighing scales

• Device that measures body weight and fat
  percentage, wirelessly transmitted to LISS
  database
• Provided to 1,000 households
• in LISS panel
• 2011 80,000 measurements
• Data can be linked to a wealth of
• socioeconomic variables

15
Weekly weight cycle
Source Kooreman and Scherpenzeel, 2012
16
Accelerometer study
Device provides aggregate indices of activity
through the day (24 hours) Wearing the device
for 8 days September 2012 feasibility pilot
among 200 LISS respondents International
collaboration American Life Panel (ALP)
English Longitudinal Study of Aging (ELSA)
17
Smartphone data collection

• Time Use Research (TUR)usually carried out using
  questionnaires and diaries?current technology
  smartphones and apps
• Also after consent, connect diary to movements
  measured by GPS

18
Pilot experiments(Fernee et al., 2012)

• Currently lt 25 of LISS respondents has
  smartphone with Internet access
• Explore the possibility to loan smartphones to
  LISS respondents
• 50 respondents with Android smartphone? download
  app from Android Play-store
• 50 respondents without Android smartphone
• ? app was already installed

19
Response
Day 1 Wednesday Smartphone owners (50) Inexperienced users (50)
Day questions uploaded (complete) 40 27
Partially complete 2 15
No observations 3 7
Never logged in 5 1
Day 2 Saturday Smartphone owners (50) Inexperienced users (50)
Day questions uploaded (complete) 37 46
Partially complete 9 1
No observations 3 0
Never logged in 1 3
20
Smartphones next steps

• September 2012 start collecting the actual TUR
  data on iPhones, Android (loan) smartphones
• (in cooperation with The Netherlands Institute
  for Social Research)
• One year each month a different batch of about
  170 respondents
• Next to the TUR study
• Project measuring travel behavior (GPS)
• Software development for smartphones

21
Linking to admin data

• Expands the possibilities for scientific research
  enormously
• Statistics Netherlands offers a remote access
  facility to (registered) researchers
• First project merging
• LISS data about minimal and adequate expenditure
  levels during retirement
• with
• administrative data on pension entitlements and
  various asset categories

22
LISS data archive and DDI

• Disseminated through website / data archive
  http//www.lissdata.nl
• Database is based on DDI 3 one of the first
  implementations of version 3(DDI Data
  Documentation Initiative,International XML-based
  standard for data documentation)
• Supports Baskets, allowing researchers to
  create and download custom datasets

23
www.lissdata.nl