This document discusses adding cell phone interviews to an existing long-term consumer sentiment study. A test was conducted adding 75 weekly cell phone interviews and 8 Spanish interviews. Results showed no significant differences in overall consumer sentiment index estimates between the traditional landline-only sample and mixed landline/cell phone sample. However, the mixed sample better represented the population demographically. Based on these findings, the document concludes it is time to fully transition the study to a mixed landline/cell phone methodology.
1.1 A Blueprint for Ending Youth Homelessness
Speaker: Eric Rice
How do we end youth homelessness? This workshop will summarize research and examine an emerging typology that can be used to inform and appropriately scale interventions to end youth homelessness. Presenters will describe strategies that are working to help young people reconnect with family and other caring adults when appropriate, and prepare to transition successfully to independent living with housing and supportive services.
Planning the Evaluation
Impact models
Types of inference and choice of design
Defining the indicators and obtaining the data
Carrying out the evaluation
Disseminating evaluation findings
Working in large-scale evaluations
So what difference does it make? Assessing the impact of participation, transparency and accountability
IDS Research Fellow, John Gaventa
World Bank Institute Seminar November 22, 2010
Four different views of a policy model: an analysis and some suggestionsBruce Edmonds
A policy model has (at least) four different interpretations: (a) intention: the intention/interpretation of the simulation designer/programmer, (b) validation: the meaning established by the validation of the model in terms of the mapping(s) to sets of evidence, (c) use: the meaning established as a result of the use of a model in a policy making/advice context and (d) interpretation: the narrative interpretation of the policy maker/advisor when justifying decisions made where this refers to a policy model.
These four different interpretations are loosely connected via social processes. The relation between intention and validation is relatively well discussed in the context of “scientific” model specification and development. The relation between use and interpretation has been discussed in a number of specific contexts. However when and how a relationship between the scientific world of intention/validation and the policy world of use/interpretation are established in practice is an area with little active research.
Both personal experience and philosophical considerations suggest that these two worlds are very different in terms of both purpose and method. However this does not mean that there cannot be any well-founded connection between them. The key question is understanding the social processes of how this can happen, what are the conditions that facilitate it happening and what is the nature of the relationship between the four views when it does happen.
Interestingly these issues have been faced and extensively discussed in the field of Artificial Intelligence, which has confronted the distinction between meaning of internal models (loosely, the beliefs of an agent about its environment) in these four ways. The field of AI has not come up with a final solution to these problems, and is itself divided into those that inhabit separate approaches that adopt a subset of these approaches to model meaning. However it is suggestive of some ways forward, namely:
• a recognition of the problem that there are these different ways of attributing meaning to a policy model (and hence avoid some common errors derived from conflating these four views);
• symbol grounding in the sense of learning meanings through repeated use and adjustment (either in response to validation or interpretation views or both);
• and the observation of scientific-policy interaction as it actually occurs (e.g. an ethnographic study of scientist/policy advisor interaction).
Some developments in the area of participatory policy modelling can be seen as forays into this arena, albeit without structured assessment.
1.1 A Blueprint for Ending Youth Homelessness
Speaker: Eric Rice
How do we end youth homelessness? This workshop will summarize research and examine an emerging typology that can be used to inform and appropriately scale interventions to end youth homelessness. Presenters will describe strategies that are working to help young people reconnect with family and other caring adults when appropriate, and prepare to transition successfully to independent living with housing and supportive services.
Planning the Evaluation
Impact models
Types of inference and choice of design
Defining the indicators and obtaining the data
Carrying out the evaluation
Disseminating evaluation findings
Working in large-scale evaluations
So what difference does it make? Assessing the impact of participation, transparency and accountability
IDS Research Fellow, John Gaventa
World Bank Institute Seminar November 22, 2010
Four different views of a policy model: an analysis and some suggestionsBruce Edmonds
A policy model has (at least) four different interpretations: (a) intention: the intention/interpretation of the simulation designer/programmer, (b) validation: the meaning established by the validation of the model in terms of the mapping(s) to sets of evidence, (c) use: the meaning established as a result of the use of a model in a policy making/advice context and (d) interpretation: the narrative interpretation of the policy maker/advisor when justifying decisions made where this refers to a policy model.
These four different interpretations are loosely connected via social processes. The relation between intention and validation is relatively well discussed in the context of “scientific” model specification and development. The relation between use and interpretation has been discussed in a number of specific contexts. However when and how a relationship between the scientific world of intention/validation and the policy world of use/interpretation are established in practice is an area with little active research.
Both personal experience and philosophical considerations suggest that these two worlds are very different in terms of both purpose and method. However this does not mean that there cannot be any well-founded connection between them. The key question is understanding the social processes of how this can happen, what are the conditions that facilitate it happening and what is the nature of the relationship between the four views when it does happen.
Interestingly these issues have been faced and extensively discussed in the field of Artificial Intelligence, which has confronted the distinction between meaning of internal models (loosely, the beliefs of an agent about its environment) in these four ways. The field of AI has not come up with a final solution to these problems, and is itself divided into those that inhabit separate approaches that adopt a subset of these approaches to model meaning. However it is suggestive of some ways forward, namely:
• a recognition of the problem that there are these different ways of attributing meaning to a policy model (and hence avoid some common errors derived from conflating these four views);
• symbol grounding in the sense of learning meanings through repeated use and adjustment (either in response to validation or interpretation views or both);
• and the observation of scientific-policy interaction as it actually occurs (e.g. an ethnographic study of scientist/policy advisor interaction).
Some developments in the area of participatory policy modelling can be seen as forays into this arena, albeit without structured assessment.
Speaker: Paul Toro
How do we end youth homelessness? This workshop will summarize research and examine an emerging typology that can be used to inform and appropriately scale interventions to end youth homelessness. Presenters will describe strategies that are working to help young people reconnect with family and other caring adults when appropriate, and prepare to transition successfully to independent living with housing and supportive services.
Engaging Youth in Project Evaluation: Why Social Media Might be the AnswerChristine Wilkinson
This is a project for my Qualitative Research Methods Course.
Youth have recently made increased their presence on social media platforms. It is imperative that project evaluation methods engage youth and encourage their participation. Social media is a great way to engage young people in project evaluation!
Summary of key findings from research by Bella Reichard and colleagues analysing high versus low scoring case studies from REF2014. View full slide deck here: https://www.slideshare.net/MarkReed11/language-in-ref2014-impact-case-studies-what-might-it-mean-for-ref2021. Read the paper here: https://www.nature.com/articles/s41599-020-0394-7
Classification and Detection of Micro-Level Impact-CSCW2017 (Link: http://dl....R R
Rezapour R, Diesner J (2017) Classification and Detection of Micro-Level Impact of Issue-Focused Films based on Reviews. Proceedings of 20th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2017), Portland, OR.
Impact Evaluation Training with AERC: China Cash Transfer Programme Technical...The Transfer Project
A hypothetical technical proposal for China's conditional cash transfer programme from our impact evaluation training with AERC in Nairobi, Kenya in July 2019.
Barry Fong, Principal Social Policy Analyst at the Greater London Authority (GLA) will take us through the Survey of Londoners 2021-22. Conducted at the end of 2021, so just before the full effects of the cost-of-living crisis began to set in, it was commissioned to provide vital evidence on key social outcomes for Londoners, following the onset of COVID-19 and associated restrictions.
A similar survey was conducted in 2018-19, so this survey would show how things had changed in the capital since then.
Barry will go through some of the key findings from the survey before handing over to Michael Cheetham and Ellen Bloomer from the North East London Integrated Care Board, who collaborated with local authority partners to fund a sample boost for the survey within North East London. They will explain how they used the data, including the analyses, the results and how this impacted strategy and practice.
International Food Policy Research Institute (IFPRI) organized a three days Training Workshop on ‘Monitoring and Evaluation Methods’ on 10-12 March 2014 in New Delhi, India. The workshop is part of an IFAD grant to IFPRI to partner in the Monitoring and Evaluation component of the ongoing projects in the region. The three day workshop is intended to be a collaborative affair between project directors, M & E leaders and M & E experts. As part of the workshop, detailed interaction will take place on the evaluation routines involving sampling, questionnaire development, data collection and management techniques and production of an evaluation report. The workshop is designed to better understand the M & E needs of various projects that are at different stages of implementation. Both the generic issues involved in M & E programs as well as project specific needs will be addressed in the workshop. The objective of the workshop is to come up with a work plan for M & E domains in the IFAD projects and determine the possibilities of collaboration between IFPRI and project leaders.
Advantages & Challenges of collecting & using longitudinal studies for research and policy.
Marta Favara, Senior Research Officer & Paul Dornan, Senior Policy Officer
Young Lives, University of Oxford
DFID Statistics Conference
6 September 2016
Speaker: Paul Toro
How do we end youth homelessness? This workshop will summarize research and examine an emerging typology that can be used to inform and appropriately scale interventions to end youth homelessness. Presenters will describe strategies that are working to help young people reconnect with family and other caring adults when appropriate, and prepare to transition successfully to independent living with housing and supportive services.
Engaging Youth in Project Evaluation: Why Social Media Might be the AnswerChristine Wilkinson
This is a project for my Qualitative Research Methods Course.
Youth have recently made increased their presence on social media platforms. It is imperative that project evaluation methods engage youth and encourage their participation. Social media is a great way to engage young people in project evaluation!
Summary of key findings from research by Bella Reichard and colleagues analysing high versus low scoring case studies from REF2014. View full slide deck here: https://www.slideshare.net/MarkReed11/language-in-ref2014-impact-case-studies-what-might-it-mean-for-ref2021. Read the paper here: https://www.nature.com/articles/s41599-020-0394-7
Classification and Detection of Micro-Level Impact-CSCW2017 (Link: http://dl....R R
Rezapour R, Diesner J (2017) Classification and Detection of Micro-Level Impact of Issue-Focused Films based on Reviews. Proceedings of 20th ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2017), Portland, OR.
Impact Evaluation Training with AERC: China Cash Transfer Programme Technical...The Transfer Project
A hypothetical technical proposal for China's conditional cash transfer programme from our impact evaluation training with AERC in Nairobi, Kenya in July 2019.
Barry Fong, Principal Social Policy Analyst at the Greater London Authority (GLA) will take us through the Survey of Londoners 2021-22. Conducted at the end of 2021, so just before the full effects of the cost-of-living crisis began to set in, it was commissioned to provide vital evidence on key social outcomes for Londoners, following the onset of COVID-19 and associated restrictions.
A similar survey was conducted in 2018-19, so this survey would show how things had changed in the capital since then.
Barry will go through some of the key findings from the survey before handing over to Michael Cheetham and Ellen Bloomer from the North East London Integrated Care Board, who collaborated with local authority partners to fund a sample boost for the survey within North East London. They will explain how they used the data, including the analyses, the results and how this impacted strategy and practice.
International Food Policy Research Institute (IFPRI) organized a three days Training Workshop on ‘Monitoring and Evaluation Methods’ on 10-12 March 2014 in New Delhi, India. The workshop is part of an IFAD grant to IFPRI to partner in the Monitoring and Evaluation component of the ongoing projects in the region. The three day workshop is intended to be a collaborative affair between project directors, M & E leaders and M & E experts. As part of the workshop, detailed interaction will take place on the evaluation routines involving sampling, questionnaire development, data collection and management techniques and production of an evaluation report. The workshop is designed to better understand the M & E needs of various projects that are at different stages of implementation. Both the generic issues involved in M & E programs as well as project specific needs will be addressed in the workshop. The objective of the workshop is to come up with a work plan for M & E domains in the IFAD projects and determine the possibilities of collaboration between IFPRI and project leaders.
Advantages & Challenges of collecting & using longitudinal studies for research and policy.
Marta Favara, Senior Research Officer & Paul Dornan, Senior Policy Officer
Young Lives, University of Oxford
DFID Statistics Conference
6 September 2016
Results of a study by researchers from Colorado WIN Partners involving residents of Imagine!’s Bob and Judy Charles SmartHome in Boulder. The researchers were looking at how the technology in the SmartHome was impacting the residents’ lives, and the data is very promising! Thanks to WIN Partners for allowing us to share this slideshow.
Greendex 2014 - Consumer Choice and the Environment - A Worldwide Tracking Su...Sustainable Brands
This report delivers additional insight specifically related to food consumption and behavior change, as well as the food-related components of the Greendex. The report seeks to better enable behavior change, given that society has not seen the pace and scale of change that is in our view required. Instead, overall Greendex scores have
remained static.
The report explores consumers’ attitudes around food consumption and production, consumers’ trust in science, choices consumers make and intend to make around food, and drivers of behavior change in this area. The report also presents a consumer segmentation based on behaviors and intentions specifically related to food.
Harold Alderman, Dan Gilligan, Melissa Hidrobo, Jessica Leight, Michael Mulford, Heleene Tambet
REGIONAL WORKSHOP
SPIR II Learning Event
Co-organized by IFPRI, USAID, CARE, ORDA, and World Vision
MAY 16, 2023 - 9:00AM TO MAY 17, 2023 - 5:00PM EAT
Topics:
Quantitative research
Characteristics of Quantitative Research
Strengths of Quantitative Research
Weaknesses of Quantitative Research
Importance of Quantitative Research Across Fields
TYPES OF QUANTITATIVE RESEARCH DESIGN
Public Attitudes toward Income Redistribution in Hong Kong April 2023.pptxstellayinying
The drivers of public support for redistributive policy have stimulated academic debate around the world. The majority of studies use cross-country surveys conducted in the Organisation for Economic Co-operation and Development countries to contribute to the debate on whether self-interest or social values have more influence on public attitudes towards redistribution. Drawing on a phone survey conducted in 2013, this study advances the discussion by investigating public attitudes towards redistribution and social policy changes against the backdrop of buoyant government revenues in Hong Kong. The Hong Kong welfare model, best seen as a parallel to the liberal welfare state, is selective and residual. Contrary to the usual assumption, the social values hypothesis, viewing poverty as societal problems instead of individual reasons, has been supported in the Hong Kong context. It lends support to greater redistribution in a residual welfare state. The policy implications of the findings are also discussed.
Similar to AAPOR 2013 Langer Research: Bloomberg CCI (20)
Correlations: Bloomberg CCI and Other IndicatorsLangerResearch
The Bloomberg ® Consumer Comfort Index ™ is a 32-year-old weekly random-sample survey of Americans’ economic attitudes. More frequent than other leading consumer measures and highly correlated on a leading basis with key economic variables, the CCI is invaluable in modeling for economists, econometricians, traders and retailers, and as a precise summary of economic attitudes for political and public policy researchers.
The 2016 Election - How and why it's President TrumpLangerResearch
Presented at AAPOR 2017 by Sofi Sinozich, Research Analyst at Langer Research Associates, and Gregory Holyk, Senior Research Analyst at Langer Research Associates
Impact Assessment: Bangladesh Leadership Development ProgramLangerResearch
Presented at the annual conference of the World Association for Public Opinion Research, Austin, TX, May 11, 2016
Nurhan Kocaoglu, Zahra Lutfeali – Counterpart International
Julie E. Phelan, Gary Langer, Gregory G. Holyk – Langer Research Associates
Matthew Warshaw – D3 Systems Inc.
The 2016 Elections: Exit Polls and TrumpismoLangerResearch
Presented at the annual meeting of the American Association for Public Opinion Research, Austin, TX, May 14, 2015
Gary Langer, Chad Kiewet de Jonge and Gregory Holyk
Langer Research Associates
Attitudes on Climate Change: Expressed Belief and Preference x PriorityLangerResearch
Presented at the annual meeting of the American Association for Public Opinion Research, Austin, TX, May 15, 2016.
Gary Langer, president
Langer Research Associates
Presented at the annual meeting of the American Association for Public Opinion Research, Austin, TX, May 13, 2016
David Cloud, Kristen Knutson - National Sleep Foundation
Julie E. Phelan, Gary Langer - Langer Research Associates
Presentation by Gary Langer on research on patient engagement and primary care redesign in California's safety-net clinics, produced from 2011-14 in partnership with Blue Shield of California Foundation, at the Clinic Leadership Institute, San Francisco, California, June 15, 2015.
Afghanistan: After the Election. Produced by the Afghan Center for Socio-economic and Opinion Research with D3 Systems and Langer Research Associates. Presented by Gregory H. Holyk at the annual meeting of the American Association for Public Opinion Research, May 16, 2015, in Hollywood, Florida.
Analysis of ABC News/Washington Post poll results on the 2014 midterm elections and a look ahead to 2016. Produced for ABC News by Langer Research Associates and presented by Gary Langer at the annual meeting of the American Association for Public Opinion Research, May 15, 2015, in Hollywood, Florida.
Opportunity Survey: Understanding the Roots of Attitudes on Inequality. Produced for the Opportunity Agenda by Langer Research Associates and presented by Gary Langer at the annual meeting of the American Association for Public Opinion Research, May 15, 2015, in Hollywood, Florida.
Presentation at the Pakistan Afghanistan Federation Forum, U.S. Pentagon, June 6, 2014, by Gary Langer, Langer Research Associates, and Matthew Warshaw, D3 Systems/ACSOR, on D3's Afghan Futures public opinion polling on the Afghan presidential election.
04062024_First India Newspaper Jaipur.pdfFIRST INDIA
Find Latest India News and Breaking News these days from India on Politics, Business, Entertainment, Technology, Sports, Lifestyle and Coronavirus News in India and the world over that you can't miss. For real time update Visit our social media handle. Read First India NewsPaper in your morning replace. Visit First India.
CLICK:- https://firstindia.co.in/
#First_India_NewsPaper
El Puerto de Algeciras continúa un año más como el más eficiente del continente europeo y vuelve a situarse en el “top ten” mundial, según el informe The Container Port Performance Index 2023 (CPPI), elaborado por el Banco Mundial y la consultora S&P Global.
El informe CPPI utiliza dos enfoques metodológicos diferentes para calcular la clasificación del índice: uno administrativo o técnico y otro estadístico, basado en análisis factorial (FA). Según los autores, esta dualidad pretende asegurar una clasificación que refleje con precisión el rendimiento real del puerto, a la vez que sea estadísticamente sólida. En esta edición del informe CPPI 2023, se han empleado los mismos enfoques metodológicos y se ha aplicado un método de agregación de clasificaciones para combinar los resultados de ambos enfoques y obtener una clasificación agregada.
‘वोटर्स विल मस्ट प्रीवेल’ (मतदाताओं को जीतना होगा) अभियान द्वारा जारी हेल्पलाइन नंबर, 4 जून को सुबह 7 बजे से दोपहर 12 बजे तक मतगणना प्रक्रिया में कहीं भी किसी भी तरह के उल्लंघन की रिपोर्ट करने के लिए खुला रहेगा।
Here is Gabe Whitley's response to my defamation lawsuit for him calling me a rapist and perjurer in court documents.
You have to read it to believe it, but after you read it, you won't believe it. And I included eight examples of defamatory statements/
An astonishing, first-of-its-kind, report by the NYT assessing damage in Ukraine. Even if the war ends tomorrow, in many places there will be nothing to go back to.
01062024_First India Newspaper Jaipur.pdfFIRST INDIA
Find Latest India News and Breaking News these days from India on Politics, Business, Entertainment, Technology, Sports, Lifestyle and Coronavirus News in India and the world over that you can't miss. For real time update Visit our social media handle. Read First India NewsPaper in your morning replace. Visit First India.
CLICK:- https://firstindia.co.in/
#First_India_NewsPaper
03062024_First India Newspaper Jaipur.pdfFIRST INDIA
Find Latest India News and Breaking News these days from India on Politics, Business, Entertainment, Technology, Sports, Lifestyle and Coronavirus News in India and the world over that you can't miss. For real time update Visit our social media handle. Read First India NewsPaper in your morning replace. Visit First India.
CLICK:- https://firstindia.co.in/
#First_India_NewsPaper
1. Adding Cell Phone Sampling to a
Long-Term Trend Study (Cautiously):
The Bloomberg Consumer Comfort Index
Julie Phelan & Gary Langer
Langer Research Associates
@LangerResearch
2. The CCI
• Continuous weekly tracking of consumer sentiment
since December 1985
• N=250/week summed in a four-week rolling avg.
• Computed as avg. pos-neg on nat’l economy, buying
climate, personal finances; -100 to +100
• Followed in financial markets and by economists and
econometricians
• Traditional design: random digit-dialed landline
telephone interviews
3.
4. Research Questions
• Should cell interviews be added to address growing
noncoverage of cell-only HHs?
• How about Spanish interviews, to better represent the
growing Hispanic population?
• Will these impact trend data, and if so how?
• Will they improve sample representativeness?
5. Operating Principles
• Any change in methodology must be undertaken
cautiously, particularly when the project relies
heavily on trend; but
• Inaction carries its own risks, to the extent that
changes in the nature or accessibility of the target
population themselves represent a de-facto
methodological departure
6. The Cell-Phone Issue
• Do LL samples adequately represent the national
population?
• Weighting to Census corrects demographic discrepancies
• Many papers show minimal diffs. in variables of interest
• But…
• Untested or growing diffs. over time are open questions
• 34% noncoverage raises face validity issues
• LL produces small sample sizes in undercovered groups
• LL has higher design effect due to weighting
7. Test Design
• LL augmented with 75 weekly cell interviews, Jan. to
March 2012, N = 2,750
• Included ~8 Spanish interviews each week, N = 82
overall (43 LL, 39 cell)
• Mixed samples produced for comparison
• Result, eight weeks of full four-week CCI estimates
8. Sample Weights
• LL sample weighted per Census as usual
• Mixed sample: weights added to account for dual-
frame respondents and adjust for phone usage in
post-stratification
• Mixed sample, English-only: weights without the
Spanish interviews
9. CCI Estimates by Sample Type
Aggregated Responses to CCI Questions Over 11-Week Period
LL sample Mixed sample
(all interviews)
Mixed sample (English
interviews)
National economy:
Positive NET 14.8% 13.4% 13.3%
Negative NET 85.2% 86.6% 86.6%
Personal finances:
Positive NET 49.9% 49.1% 49.8%
Negative NET 50.1% 50.9% 50.2%
Buying climate:
Positive NET 29.3% 28.1% 28.5%
Negative NET 70.7% 71.9% 71.5%
CCI -37.3 -39.6 -38.9
Differences in aggregated CCI:
• -1.6 with inclusion of cell phones
• -0.7 with inclusion of Spanish interviews
• No differences are statistically significant
11. Weekly Ratings by Sample
Positive Ratings by Sample and Week
National economy Personal finances Buying climate
LL Mixed Sig? LL Mixed Sig? LL Mixed Sig?
2/5/12 12% 11% N 50% 48% N 26% 27% N
2/12/12 14% 13% N 50% 48% N 26% 27% N
2/19/12 14% 13% N 51% 49% N 29% 27% N
2/26/12 14% 13% N 50% 47% N 28% 28% N
3/4/12 15% 13% N 50% 48% N 30% 27% N
3/11/12 17% 15% N 50% 50% N 33% 31% N
3/18/12 18% 15% N 49% 51% N 31% 30% N
3/25/12 17% 16% N 50% 52% N 32% 29% N
12. Summary: Weekly Ratings by Sample
• No statistically significant differences in any of the
three questions in any week
• Week-to-week variability within each sample type
exceeded average discrepancy between sample types
– Within-sample standard deviations:
• LL = 2.8 points
• Mixed = 2.9 points
– Avg. difference between sample types: 2.1 points
13. Results Among Groups
• Avg. differences by sample type among
demographic groups were greater than the
difference between the full samples (not
surprising given smaller sample sizes).
• But differences were not unidirectional:
– Mixed sample numerically higher in 12 of 36 cases
– LL sample numerically higher in 24 of 36 cases
• In 31 of 36 groups examined, avg. difference
between sample types was smaller than the avg.
variability within each sample type.
14. Groups with Differences
• One was N <100 in LL sample (age 18-34)
• Other four:
– $100,000+ income: 10.2 pt difference
– Westerners: 9.2 pt difference
– High school graduates: 7.2 pt difference
– Political independents: 6.2 pt difference
• We conducted z-tests on each of the three CCI questions to
test whether the differences were statistically significant
– 96 comparisons: 3 questions x 4 response options x 8 weeks
– Just one statistically significant difference: On 3/18/12,
significantly more independents rated the national economy
positively in the LL sample vs. the mixed sample
15. Results of CCI Comparison
• Differences between LL and mixed samples
overall were trivial:
– Responses to the three index questions did not
meaningfully differ by sample type, either across
the full period or by week.
– Overall CCI estimates differed very little, and
results followed the same trajectory.
• Variation by sample type among groups was
no more than would be expected by random
chance.
16. Next Question: Sample Quality
• Absence of a discernible impact of cells and Spanish
can support a decision to switch or not
• What re growing noncoverage of groups most likely
to lack landline service (e.g., young, renters,
minorities, lower-income)?
• We compared the unweighted demographic
composition of the two samples
17. Comparing Sample Quality
• Result: We find statistically significant differences
between the unweighted LL and mixed samples in 13
of 38 variables studied:
– Sex, age (18-34, 55-64, and 65+), race (white and
Hispanic), home ownership, relationship status
(never married and widowed), employment (full-
time and not at all)
• In every case, the mixed frame sample was closer to
Census norms
18. Differences Between Unweighted
Estimates and Census CPS Targets
Benchmarks LL sample Mixed sample
Men 48.6% 43.8% 48.6%
Women 51.4% 56.2% 51.4%
Age: 18-34 30.5% 9.1% 17.1%
Age: 55-64 15.8% 24.4% 21.4%
Age: 65+ 16.7% 35.3% 29.1%
White 66.9% 80.1% 75.0%
Hispanic 13.9% 4.9% 9.1%
Home: Owned 68.8% 79.8% 73.2%
Home: Rented 30.2% 19.3% 25.5%
Never married 27.1% 15.2% 21.2%
Widowed 6.1% 14.9% 12.2%
Not employed 40.1% 52.1% 46.7%
Average absolute differences:
• Benchmarks vs. LL: 11.25 pts
• Benchmarks vs. mixed: 6 pts
19. Sample Quality Summary
• The unweighted sample obtained using the mixed-
frame design is far superior to the sample obtained
using the LL design
– In every case in which demographic estimates differed by
sample type, the mixed sample estimate more closely
matched the true population parameters
– The mixed design provided greater coverage of groups
including young adults, men, lower-income respondents,
non-whites and renters
20. Design Effect
• Design effect = estimate of the impact on
sampling error caused by a survey’s departure
from simple random sampling.
– The larger the weights needed to correct for non-
response and noncoverage, the higher the deff.
– Larger design effects = more difficult to find
statistically significant differences.
• LL deff = 2.3 (MoE for N=1000 is +/- 4.7 pts.)
• Mixed deff = 1.5 (MoE for N=1000 is +/- 3.8 pts.)
21. Conclusion
• No material, consistent impact of including cell-phones
on the Bloomberg CCI.
– week-to-week pattern of results was similar.
– differences by sample type were less than normal weekly
variation.
• Few differences among subgroups, and these may reflect
chance variation, since the CCI among groups typically
differed more within samples than between sample types.
• Better coverage, lower design effect/lower MoE and
greater face validity with no disruption to trend.
• Time to make the switch!
Editor's Notes
No differences are statistically significant. Tested using a z-test that accounts for the overlap between the samples (the more conservative approach because adjusting for the overlap makes it easier to find statistically significant differences)
These and all subsequent analyses compared the full mixed sample (including Spanish interviews) to the LL sample. We used the full mixed sample for two reasons: one, comparisons involving the full mixed sample (hereafter referred to simply as the “mixed sample”) provide a more stringent test of the impact of the sampling shift and two, the Spanish-inclusive estimates are preferable on theoretical grounds.As noted, some of the difference between the full mixed-sample CCI and the LL CCI was due to the inclusion of Spanish language interviews. Therefore if analyses show no differences between the full mixed sample and the landline sample we can be sure there are no differences between the English-only mixed sample and the LL sample.
Each of the differences in unweighted estimates between the LL and mixed samples shown in this table are statistically significant, and in every case the unweighted mixed sample more closely matches the census/cps targets.For the sake of simplicity, rather than having separate slightly different targets for each of the samples (which is how I did it in the paper in order to adjust for missing data), I’m just listing one target (averaging the two). I don’t think anyone will say anything, but if they ask why this table is slightly different than the numbers in Tables 5 and 6 in the paper, that’s why.