Agenda

1 / 27

# Agenda - PowerPoint PPT Presentation

Agenda. Recap: Sample design and response rates Historical and comparative research Using archival resources Group research proposals. Respondent. Age. Mean:. n i= 1. . x i. 1 2 3 4 5 6 7 8 9 10. 20 22 22 23 24 24 24 25 28 30. (20+22+22+23 … +30). 242. x =. =.

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.

## PowerPoint Slideshow about 'Agenda' - nora-franco

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Agenda
• Recap: Sample design and response rates
• Historical and comparative research
• Using archival resources
• Group research proposals

Respondent

Age

Mean:

n

i=1

xi

1

2

3

4

5

6

7

8

9

10

20

22

22

23

24

24

24

25

28

30

(20+22+22+23 … +30)

242

x =

=

=

=

24.2

N

10

10

Variance:

n

i=1

(xi – x)2

s2=

=

(20-24.2)2 + (22-24.2)2 …

77.6

=

=

8.62

N-1

10 - 1

9

Standard Deviation:

__

s = s2

=

8.62

=

2.94

Standard Error:

s

2.94

s.e. =

___

=

=

.93

N

10

95% Confidence Interval:

24.2 +/- (1.96 x .93) = 24.2 +/- 1.82 = 22.4 to 26.0

Sampling design
• Most samples not simple random
• RDD telephone surveys weighted for number of lines per household
• Cluster samples adjusted for number of units sampled at each stage

Primary Sampling Unit (PSU)

e.g., SMA

Second-Stage Sampling Unit (SSU)

e.g., area segments (Census blocks)

Housing Unit

As the number of PSUs and SSUs declines, sample variances and sampling errors will increase

The Design Effect is the ratio of the variance of a particular sample to a simple random sample of the same size

A design effect < 1 is more efficient than simple random

A design effect > 1 is less efficient than simple random

Selected Respondent

Survey nonresponse

CONSEQUENCES OF REDUCING NONRESPONSE IN A NATIONAL TELEPHONE SURVEY

SCOTT KEETER

CAROLYN MILLER

ANDREW KOHUT

ROBERT M. GROVES

STANLEY PRESSER

AbstractCritics of public opinion polls often claim that methodological shortcuts taken to collect timely data produce biased results. This study compares two random digit dial national telephone surveys that used identical questionnaires but very different levels of effort: a "Standard" survey conducted over a 5-day period that used a sample of adults who were home when the interviewer called, and a "Rigorous“ survey conducted over an 8-week period that used random selection from among all adult household members. Response rates, computed according to AAPOR guidelines, were 60.6 percent for the Rigorous and 36.0 percent for the Standard study. Nonetheless, the two surveys produced similar results. Across 91 comparisons, no difference exceeded 9 percentage points, and the average difference was about 2 percentage points. Most of the statistically significant differences were among demographic items. Very few significant differences were found on attention to media and engagement in politics, social trust and connectedness, and most social and political attitudes, including even those toward surveys.

Public Opinion Quarterly, Summer 2000, pp. 125-148

Historical and comparative research
• Historical Events Research
• Historical Process Research
• Cross-sectional Comparative Research
• Comparative Historical Research

Cross-Sectional Longitudinal

Historical

Events

Research

Historical Process Research

Single

Case

Multiple

Cases

Cross-sectional Comparative Research

Comparative Historical Research

Examples

Robert Putnam’s research on social capital

• Comparative studies of Italian regional governments
• Historical trends in American social capital
• Comparative studies of U.S. States
Social capital
• Features of social life that enable people to act together more effectively
• Social networks
• Social trust
• Social norms
• Accrued collectively, like financial capital
• Theoretically central to democratic life
Comparative study
• Fifteen Italian regional governments
• Examined institutional success
• Stability of cabinets
• Prompt budgets
• Legislative innovations
• Predictors of success
• Economic development
• Social stability
• Civic culture (voluntary associations)
Historical study
• U.S. trends in social group memberships and social trust
• 25% decline in group memberships since 1972
• 30% decline in interpersonal trust since 1972
• Finds declines in political participation as well

Group

Member-

ships

2

Proportion

Saying

People Can

be Trusted

1

52%

Long civic generation

25%

1900

1940

1970

Year of Birth

Why the declines?

A “mysterious anti-civic X-ray” –

television

• The timing fits (temporal order)
• TV viewing associated with lower social capital (correlation)
• Plausible mechanisms to explain effect
• Time displacement
• Cultivation of outlooks
• Childhood socialization
Alternative explanations

Putnam examines several:

• Pressures of time and money
• Mobility and suburbanization
• Changing role of women
• Marriage and family
• Rise of welfare state
Counter-arguments
• Not declines but transformations of traditional forms of civic-mindedness
• Growth of school and health-related activities
• Timing doesn’t quite fit a TV explanation
• Declines register in the 1950s -- among people who predate the TV generation

Turned 18 in late 1940s

and 1950s

Group

Member-

ships

2

Proportion

Saying

People Can

be Trusted

1

52%

25%

1900

1940

1970

Year of Birth

Methodological issues
• Reliance upon archival data sources
• Secondary analysis
• Measurement across contexts
Archival data
• The “running record” may be non-reactive
• Data gathering independent of the researcher
• How gathered? For what purpose?
• Selective deposit and selective survival
• Consider Congressional Record, archeological use of pottery
Secondary analysis

Putting data to purposes for which they were not originally gathered

• Fit between theoretical constructs and available measures may be rough
• Key analytic variables may be missing
Comparability of data
• Comparisons across countries, cities, etc.
• Differences in sampling
• Differences in language (e.g., comparing “trust”)
• Political and cultural differences (e.g., comparing voluntary associations)
• Comparisons over time
• Change over time in meanings (e.g., comparing racial attitudes)
• Differences in samples, questions, and survey context
Archival resources
• Largest single archive of social science data in the world
• Data and documentation provided free to member institutions
Archival resources
• Decennial Census
• Monthly Current Population Survey (CPS)