INFORMING CONTINGENT VALUATION METHODS WITH INTERNET SURVEYS - PowerPoint PPT Presentation

1 / 33
About This Presentation
Title:

INFORMING CONTINGENT VALUATION METHODS WITH INTERNET SURVEYS

Description:

State of Ohio v. Department of Interior (1989) ... Respondent demographics, politics, and computer use. BASIC KYOTO PROTOCOL QUESTION ... – PowerPoint PPT presentation

Number of Views:32
Avg rating:3.0/5.0
Slides: 34
Provided by: joannabals
Category:

less

Transcript and Presenter's Notes

Title: INFORMING CONTINGENT VALUATION METHODS WITH INTERNET SURVEYS


1
INFORMING CONTINGENT VALUATION METHODS WITH
INTERNET SURVEYS
  • Robert P. Berrens, Alok K. Bohara, Carol Silva,
    Hank Jenkins-Smith, Dave Weimer
  • Project funded by the Decision, Risk and
    Management Science Program, National Science
    Foundation http//www.nsf.gov/cgi-bin/showaward?a
    ward9818108

2
OVERVIEW
  • Contingent valuation surveys and passive use
    benefits
  • Decreasing viability of telephone surveys?
  • Increasing viability of Internet surveys?
  • Study design telephone and Internet splits
  • Mental accounts (telephone and Internet)
  • Enhanced information (Internet)
  • Modified Kyoto Protocol (Internet)
  • Preliminary Findings

3
GENERAL QUESTIONS
  • Technological question
  • Can the Internet be used as an alternative to
    the telephone for administering contingent
    valuation (CV) surveys?
  • If yes, then dramatic revolution!
  • If no, then how far off?

4
GENERAL QUESTIONS, Continued
  • Methodological questions Whether or not the
    Internet is a survey replacement technology for
    the telephone, can the Internet be used as a
    low-cost tool for investigating methodological
    issues about CV?
  • Role of greatly increased availability of
    information
  • Role of mental account questions

5
GENERAL QUESTIONS, Continued
  • Substantive questions
  • What is the willingness of the U.S. population to
    pay for ratification of the Kyoto Protocol?
  • Would willingness to pay be substantially higher
    if developing countries had emission limits?

6
CONTINGENT VALUATION IS HERE TO STAY
  • Desire to include passive-use benefits in
    cost-benefit analysis
  • Direct correspondence to option price as benefit
    measure
  • "Structured conversation" when no "behavioral
    traces"
  • State of Ohio v. Department of Interior (1989)
  • Passive-use values valid in CERCLA damage
    assessment
  • Oil Pollution Act of 1990
  • Increased use of CV in natural resource damage
    litigation

7
GROWING LEGITIMACY
  • NOAA Panel (Arrow et al., 1993)
  • Qualified endorsement by blue ribbon panel
  • Academic mainstream (environmental economists)
  • Since January 1990 the Journal of Environmental
    Economics and Management and Land Economics have
    published 65 articles reporting on CV surveys
    (mean sample size is 730)

8
CV METHODOLOGICAL ISSUES
  • Elicitation method
  • Dichotomous choice/policy referendum format
    preferred
  • Bishop and Heberlein (1979)
  • NOAA Panel recommends
  • Incentive compatibility (Carson, Groves, Machina,
    1999)
  • Consequential questions potential effect,
    matters
  • Public good with coercive payments most likely
    incentive compatible
  • Payment vehicle
  • Non-commitment bias reminder needed
  • "Top down disaggregation method" or mental
    accounts

9
MORE ISSUES
  • Description of good
  • "Hypotheticality problems"
  • Meaning and context must be understood
  • Decision making
  • Embedding effects responses not always scale
    dependent
  • "Warm glow broad moral attitudes rather than
    valuation
  • Willingness-to-pay versus willingness-to-accept

10
DECREASING VIABILITY OF TELEPHONE SURVEYS?
  • Random digit dialing as "probability sample" of
    households
  • Decreasing response rates, or increasing cost per
    response
  • Hectic pace of life
  • Increased telemarketing, sometimes in guise of
    survey
  • Skepticism related to perceived biases
  • Increasing complexity of telephone system
  • More lines per household
  • Lines dedicated to facsimile machines and
    computers
  • More cellular telephones (86 million subscribers
    in 1999)
  • More answering machines more call screening
  • Breakdown of geographic area codes
  • More non-working numbers (Probability of reaching
    working number fell from 21 in 1988 to 13 in
    1998)

11
INCREASING VIABILITY OF INTERNET SURVEYS?
Spaming prohibitions prevent any analog to RDD,
but ...
  • Creation of proprietary panels of cooperative
    respondents (non-systematic recruitment)
  • Harris Interactive panel of U.S. adults
  • 4.4 million in January 2000
  • 4.8 million in August 2000
  • Over 6 million now, and growing
  • Creation of proprietary panels of cooperative
    respondents (random recruitment)
  • Knowledge Networks panel of U.S. households
  • Approximately 100,000 households, growing to
    250,000
  • Provides fee equipment and Internet service
    weekly surveys

12
INTERNET PENETRATION
  • Fraction of U.S. households with Internet access
    is approaching 50
  • Women and men now use at same rate
  • Women 17 in 1997 49 in 2000
  • Ethnic minorities converging
  • African-American 33 Asian-American 69
  • Hispanic-Americans 47 Caucasian-Americans 43
  • Age profile converging rapidly, but those over 55
    still underrepresented
  • Income converging, but
  • Under 20K, 6 of Internet users, 19 of
    population
  • Over 150K, 8 of Internet users, 4 of population

13
POTENTIAL ADVANTAGES OF INTERNET SURVEYS
  • Lower cost per respondent
  • Our telephone survey about 50K for 1,700
    respondents
  • Our Internet survey about 40K for 13,000
    respondents
  • Lower cost makes larger sample sizes feasible
  • Within-versus across-study methodological
    comparisons
  • Greater sub-sample opportunities
  • Eliminates interviewer bias
  • Identification of rare respondents (medical
    research)
  • Speed (likely application in polling Harris did
    well in 2000 election forecasts)
  • Opportunity to provide more and different
    information than possible in telephone surveys

14
INTERNET SKEPTICS
  • "... The panel used as a frame to sample
    respondents is not a probability sample of
    internet users. Even if it were, internet users
    are not like non-internet users. For instance,
    they are better educated, more affluent, and more
    male. The important characteristics for weighting
    have not yet been identified. Generalizations
    from a census of the growing millions in the
    panel do not even represent internet users. The
    whole panel represents nothing but itself ... I
    can see no valid survey purpose to the current
    internet enterprise. All that will happen will be
    the accumulation of thousands upon thousands of
    interviews of dubious merit that will mislead the
    public and destroy whatever credibility surveys
    and polls now have. A growing number of survey
    researchers are unfortunately being led to the
    rocks like Ulysses sailors following the Siren
    call of cheap, but worthless, data." (pp 25-26)
  • Warren J. Mitofsky, "Pollsters.com," Public
    Perspectives, June/July 1999, 24-26.

15
DOES WEIGHTING OFFER HOPE?
  • Exxon Valdez CV (Carson et al., 1993)
  • Gold-plated, "litigation quality" survey
  • Relatively large sample 1,043
  • Cost approximately 3 million
  • Estimated damage from Exxon Valdez spill 2.128
    billion
  • "Replication" (Harrison and Lesley, 1996)
  • Convenience sample
  • North and South Carolina Students
  • Modest sample size 449
  • Simplified questionnaire format
  • Cost approximately 2,500 plus some beers
  • Estimated 4.044 billion in damages with raw
    sample data
  • Estimated 2.807 billion in damages with
    demographic weighting that applies population
    means to coefficients estimated in
    willingness-to-pay function

16
OUR STUDY
  • Telephone survey
  • Institute for Public Policy, University of New
    Mexico, January 2000 1,699 completed surveys
  • Internet survey (planned)
  • Harris Interactive, Rochester, January 2000
    13,034 completed surveys
  • Internet survey (subsequent)
  • Harris Interactive, August 2000 11,160 completed
    surveys
  • Knowledge Networks, Stanford, November 2000
    1,500 expected
  • To take survey
  • http//www.unm.edu/instpp/gcc/

17
EXPERIMENTAL DESIGN
18
SURVEY OVERVIEW
  • Demographics (education and sex)
  • Environmental attitudes
  • Global climate change knowledge
  • Split 1
  • Basic reminder/Mental accounts
  • Split 2 (Internet only)
  • Basic information/Enhanced information
  • Kyoto Protocol background
  • Split 3 (Internet only)
  • Basic Kyoto referendum/Modified Kyoto
    referendum
  • Attitudes toward appropriateness of
    willingness-to-pay
  • Perceived efficacy of Kyoto Protocol
  • Balance and usefulness of enhanced information
    (as relevant)
  • Presidential candidate questions
  • Respondent demographics, politics, and computer
    use

19
BASIC KYOTO PROTOCOL QUESTION
  • Seven preparatory questions to describe Protocol,
    then
  • The US Senate has not yet voted on whether to
    ratify the Kyoto Protocol. If the US does not
    ratify the treaty, it is very unlikely that the
    Protocol can be successfully implemented.
  • Suppose that a national vote or referendum were
    held today in which US residents could vote to
    advise their Senators whether to support or
    oppose ratifying the Kyoto Protocol. If US
    compliance with the treaty would cost your
    household X dollars per year in increased energy
    and gasoline prices, would you vote for or
    against having your Senators support ratification
    of the Kyoto Protocol? Keep in mind that the X
    dollars spent on increased energy and gasoline
    prices could not be spent on other things, such
    as other household expenses, charities,
    groceries, or car payments.
  • X is randomly chosen from 6 12 25 75 150 225
    300 500 700 900
  • (1200 and 1800 added after telephone pretest)

20
MODIFIED KYOTO PROTOCOL QUESTION
  • (Same lead-up questions)
  • The US Senate has not yet voted on whether to
    ratify the Kyoto Protocol. If the US does not
    ratify the treaty, it is very unlikely that the
    Protocol can be successfully implemented.
  • An alternative to the Kyoto Protocol, which we
    will refer to as the Modified Kyoto Protocol,
    would make only one change in the agreement It
    would require that developing countries, such as
    China, India, Mexico, Brazil, and Argentina,
    promise to restrict their future production of
    greenhouse gases to no more than 5 percent above
    current levels.

21
MENTAL ACCOUNTS TREATMENT
  • Standard reminder as part of willingness-to-pay
    question
  • Mental accounts approach
  • Questions to encourage budget consideration
  • Ask prior to valuation question
  • May be useful in estimating willingness-to-pay
    models

22
MENTAL ACCOUNTS QUESTIONS
  • First level compartment
  • Now think about your average monthly income and
    expenses.
  • After you have paid all the necessary bills for
    such things as housing, transportation,
    groceries, insurance, debt, and taxes, what
    percent of your income is left over for optional
    uses on things like recreation, savings, and
    giving for charity and other causes?
  • Second level compartment
  • Now think about the portion of your total income
    available for optional uses. On average, what
    percent of that amount do you use for
    contributions to environmental causes, such as
    donations for specific programs or contributions
    and memberships to environmental advocacy groups?
  • Note Please remember that we are not asking for
    money for anything, and that your answers will be
    kept completely confidential.

23
ENCHANCE INFORMATION
  • Information menu
  • Open through referendum question and follow-ups
  • Lists 27 entries
  • Each entry contains one page of information
  • Follow-up questions
  • Visits and time spent
  • Perceived usefulness and bias

24
INVITATION TO MENU
  • Thank you for your answers to the previous
    section. Next we would like your views on a
    particular environmental treaty, the Kyoto
    Protocol. The Kyoto Protocol is concerned with
    addressing the issue of possible human causes of
    global climate change.
  • Before proceeding to the questions on the Kyoto
    Protocol, you are invited to explore a brief
    information menu we have made available.
  • The items in the information menu are intended to
    provide brief and balanced summaries of research
    and informed opinion about global climate change,
    and the Kyoto Protocol. Because the volume of
    information on these matters is enormous and
    growing rapidly, we sought to make the items in
    the information menu broadly representative and
    readily understood, rather than exhaustive.
  • Please feel free to visit as many pages as you
    wish as you proceed through the survey questions.

25
INFORMATION MENU
  • Global Warming Theory
  • Basic "Greenhouse" Mechanism
  • Greenhouse Gases
  • Carbon Dioxide Changes
  • Feedback Mechanisms
  • Feedbacks Related to Carbon Dioxide and Methane
  • Feedbacks Related to Water Vapor, Clouds, and Ice
    Cover
  • Feedbacks Related to Ocean Currents
  • Aerosol Offsets
  • Temperature Predictions under Current Policy
  • Evidence of Global Warming
  • Ground-Based Measurements
  • Atmospheric Measurements

26
INFORMATION MENU, Continued
  • Consequences of Global Warming
  • Sea Level Changes
  • Ecological Effects
  • Agricultural Effects
  • Health Effects
  • Weather Extremes
  • Alternative Theories and Explanations for
    Apparent Warming
  • Solar Cycles
  • Measurement Issues
  • Modeling Limitations

27
INFORMATION MENU, Continued
  • Kyoto Protocol on Climate Change
  • Background
  • Country Commitments
  • Flexibility Mechanisms
  • Predicted Impacts under Full Implementation
  • Ratification Status, Political Views, and Public
    Opinion
  • Ratification Status
  • Views in Opposition to Ratification of the Kyoto
    Protocol
  • Views in Favor of Ratification of the Kyoto
    Protocol
  • Selected Reports from Public Opinion Surveys

28
Very Preliminary Findings
  • Demographics
  • Knowledge about effects and causes of global
    climate change
  • Relationships across modes
  • Use and perception of bias in information

29
Demographics
Telephone January 2000 (Semi-Weighted) Internet January 2000 (Weighted) Internet August 2000 (Weighted)
Mean Age (years) 41.8 43.7 44.4
Percent Male 47.1 48.0 48.0
Percent Some College or More 66.2 48.0 49.5
Percent Hispanic 7.5 10.0 10.2
Percent African-American 7.3 12.4 11.5
Mean Annual Household Income () 58,404 45,064 52,189
30
Knowledge Telephone Internet 1
Internet 2
31
Relationship between Ideology and Belief about
Environmental Threat (0 no real threat 10
brink of collapse)
N23,273 R2.11 Coefficient Standard Error P-Value
Ideology (1 SL 7SC) -.34 .04 .000
Ideology Internet 1 -.15 .06 .015
Ideology Internet 2 -.15 .05 .004
Internet 1 .67 .26 .010
Internet 2 .55 .24 .020
Constant 7.12 .18 .000
32
Relationship between Ideology and Belief about
Environmental Threat Controlling with Demographic
Covariates
N19,292 R2.13 Coefficient Standard Error P-Value
Ideology -.33 .05 .000
Ideology Internet 1 -.08 .06 .157
IdeologyInternet 2 -.10 .06 .089
Internet 1 .37 .25 .132
Internet 2 .28 .26 .282
Five Demographic Controls Income (-), Age, Sex (M-), College,African-American ------ ----- -----
Constant 7.70 .24 .000
33
Use and Perception of Enhanced Information
  • How many pages did you visit?
  • Mean 6.3 (Range 0 to 27)
  • How much time did you spend?
  • Mean 9.1 minutes (12.3 for visitors)
  • How useful did you find the information?
  • Mean 5.3 (0not at all 10extremely useful)
  • How balanced did you find the information?
  • Mean 5.3 (0biased against GCC 10 for)
  • Tri-modal distribution 54 at 5 10 at 0, 7 at
    10
Write a Comment
User Comments (0)
About PowerShow.com