complex survey samples explaining the miracle: statistics and analysis in public health apheo...

41
Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department of Public Health Sciences, University of Toronto

Upload: eustace-merritt

Post on 27-Dec-2015

221 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

Complex Survey Samples

Explaining the Miracle: Statistics and Analysis in Public Health

APHEO Conference 2007, October 14-16, 2007

Susan Bondy,Department of Public Health Sciences,University of Toronto

Page 2: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

2

Outline

• Goals of complex survey analysis

• What is simple, what is complex– Issues and implications of complexities

• Working with software

• Tips for working with expert analysts

Page 3: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

3

What we report from surveys

• Descriptive statistics– Mean, median, counts, totals

• Measures of difference, association and effect– % diff, risk diff, OR, RR, rho, etc.

• Always reported with expression of variance– Margin of Error (MOE or +/- part)– Confidence intervals

– Point estimate versus variance

Page 4: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

4

Meet two users of survey data

The Describer The Modeller

Page 5: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

5

The describer

• Population inference is #1ALWAYS need true pop’n rep.

samples

• Sometimes just descriptive statistics (rates)

• Interest in comparisons:– monitoring and surveillance

(e.g., across time, space, sub-populations)

– Consistency is important

The modeller

• Hypothesis tests are #1

• Analyses simulate controlled experimentsRarely need true pop’n rep.

samples

• Interest in comparison:– Replication of experiments– Differences between studies

more interestingExtending and testing theory

Page 6: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

Complex samples

Page 7: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

7

Simple Random Sample

• Selection into sample is entirely at random

• Each member of pop has same chance of being in the sample

• No strata, no clusters, self-weighting

• Statistically efficient (all observations are independent – tightest margins of error)

Page 8: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

8

Complex designs

1. Selection by cluster

2. Stratification

3. Probability sample weights

4. Finite population correction

• Worst of all:– Mishmashes of all the above– & where you can’t have the information

Page 9: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

Cluster sampling

Page 10: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

10

Cluster sampling• E.g., people by FAMILY, students by CLASS, teeth by

MOUTH , etc.,

• Now WELL recognized as a problem– Non-independence means loss of statistical power (variance

understated, if ignored)

• Need:– New statistics textbooks– More expensive software

…will return to software options

Page 11: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

11

Sample logistic results

Model-based

95%CI

Linearized 95%CI

DEFF

Sex

Grade

Region 2

Region 3

Region 4

1.6

1.5

1.4

1.3

1.2

( 1.4 - 1.8 )

( 1.4 - 1.5 )

( 0.9 - 1.7 )

( 1.1 - 1.7 )

( 0.9 - 1.5 )

( 1.3 - 2.0 )

( 1.4 - 1.5 )

( 0.9 - 2.1 )

( 0.9 - 1.9 )

( 0.8 - 1.9 )

1.4

1.7

1.5

1.9

1.8

Page 12: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

12

Repeat after me:

“Failure to account for non-independence of observations, in the analysis, will always result in an underestimation of variances”

• Confidence intervals narrower…• p-values smaller… • results ‘less conservative’ …

… than they should be

Page 13: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

Stratification

Page 14: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

14

What is: stratification?

• Division of the target population into groups or layers from which samples are drawn

• e.g., Plan for reports on– Youth – Smaller pop’n regions

Page 15: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

15

Goals of stratification

1. For PLANNED descriptions of sub-populations• E.g., regions, age-groups

2. For design correction:• To prevent extreme unrepresentativeness• e.g., empty groups; extreme weights

3. To improve precision of the overall (or full pop) estimates

Implications…

Page 16: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

16

Stratification WEIGHTS

They come as a pair

Page 17: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

17

Impact of weights in analysis

• Impacts precision – a huge DEFF issue

• Other model problems– E.g., can create highly influential observations

• Restricts software and analysis choices

When, why of weights?

Page 18: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

18

Repeat after me:

“You knew clustering affected variance estimates and had to be taken into account…

Sometimes WEIGHTS have an even bigger bad effect on precision !”

Always use software and procedures specific to complex survey data, even when weighting is your only complexity.

Page 19: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

But wait a minute, I’ve been told unweighted is sometimes better

Page 20: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

20

Scenario A

People up-weighted People down-weighted

Weighted or unweighted is same slope !

Page 21: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

21

Scenario BSomething correlated with relative weights is associated with a different slope

Low educ.

Over educated

Exposure to materials

Rea

dine

ss to

qui

t

Weighted

Page 22: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

22

Scenario C

Distance from airport (km)

Annoyance ratings (%)

Weighted slope

Unweighted slope

Page 23: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

23

Scenario C

Distance from airport (km)

Annoyance ratings (%)

Weighted or unweighted curve

Page 24: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

24

Modeller’s adage

• If weighted and unweighted differ then, both are wrong

• There must be a complex relationship, or better model, to find and describe

Page 25: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

25

Pub. Hlth. Epis. are always DESCRIBERS

Page 26: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

26

Scenario BSomething correlated with relative weights is associated with a different slope

Low educ.

Over educated

Exposure to materials

Rea

dine

ss to

qui

t

Pop’n weighted is TRUE population estimate of ‘net’ or ‘average’ effect

Page 27: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

27

Model all possible interactions with age, sex and geography strata?

Yes, – Do look for effect modification where there are good

grounds (show net and specific data)

No, – In hundreds of age*sex*region strata, some random

variation by chance – In large samples lots of meaningless interactions can

be detected

– Pop average effect is still pop average effect

Page 28: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

28

Message so far…

Can never ignore:– Cluster sampling– Weighting

So, HOW to analyze data?

Page 29: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

29

2 most commonly used for complex survey variance estimation

“Taylor-Series”aka

“Linearized” variance estimation

“Bootstrap”

Usually achieved using bootstrap

replicate resampling weights

Page 30: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

30

Taylor SeriesComplex linear equations to estimate

corrected variance for every estimate• Requires assumptions about data !

–Normally distribution assumptions –Large sample sizes

• Very difficult for user to know:–when limits are being pushed–When procedure is accepted or controversial

• Requires full design information

• Even more ‘approximate’ with more complex designs

Page 31: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

31

Using “Taylor-series” type software

1) Use syntax (or even boxes) to declare the following:

• Weight variable• Stratification variable• Group unit for cluster sampling

– Primary sampling unit or PSU• (Ignore requests for finite population info)

2) Run your analysis as available in software• Using only ‘special’ commands for complex

samples

Page 32: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

32

Survey estimates

• Prevalence = 13.0 (95% CI = 10.0-16.0)

• Odds ratio = 2.1 (95% CI = 1.6-4.0)

Usual weighted point estimate Variance

calculated from a formula;

substituted in things like CIs

Page 33: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

33

Bootstrap variance weights• Sampling variability “observed” not calculated from a fixed formula

– Felt to reflect “true” sampling variability, – As due to chance alone if survey really repeated an infinite number of times

• Virtually free of assumptions– Tends to be more appropriate and conservative when assumptions for linearization fails

• Very broadly applicable

Page 34: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

34

Creation of BRR weights• Someone takes a lot of random COMPLEX sub-

samples of the full survey dataset (~500 times)

• The full algorithm for pop’n weighting is applied to each sub-sample– When obs not in sample, weight=zero– Rest re-weighted to reflect pop’n again

• RESULT– 500 weights, – When applied to full dataset, simulates taking 500 samples

again

Page 35: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

35

Bootstrapping (with weights)

• Point estimates taken from full sample– Mean = 13.0

• Same point estimate taken from 500 B.S. samples

• Observed variability in 500 B.S. estimates becomes variance for mean of 13.0.

Page 36: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

36

Survey estimates

• Prevalence = 13.0 (95% CI = 10.0-16.0)

• Odds ratio = 2.1 (95% CI = 1.6-4.0)

Usual weighted point estimate Variance reflects

OBSERVED variance in 500

estimates of prev. and OR.

Page 37: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

37

Software options (more?)Epi Info Linearized estimation only

Limited analysis options

SPSS Linearized estimation

Several analyses available

Stata Linearized or BS Weights

Good range of ‘canned’ complex analyses

SAS Linearized

Means, prop. linear and logistic (more in v10)

Wesvar Linearized or BS weights

Statistics Canada Bootvar

BS Weights,

Bonus output: CV and suppression rules

Somewhat limited analysis options

Page 38: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

38

Beware

• Stick to procedures custom-designed for complex survey samples– Will handle weights properly– Will give useful statistics, such as DEFF

• Bootstrapping without a set of BS weights– If you aren’t screaming in pain, you haven’t

got it right

Page 39: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

39

Tips for working in partnership

1. Get a geek to generate lots of useful sets of BS Weights for your survey

• e.g., your favourite standard pop’n• Does take expertise, but done once benefits many

many users

2. Get a nerd to do only your variance corrections for you

• Use your favourite software and keep very detailed programs (recodes, restrictions, etc)

• Have them repeat very defined results tables

Page 40: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

40

Table 4 Estimated precision of estimates resulting from an overall sample of 1000 residents from each of two strata of one Health Unit. Fictional smoking survey.

Percent daily smokers

Percent (95% CI) Number of cigs/day

Mean (95% CI) Whole sample Daily smokers only All ages 15-24 25+ All ages 15-24 25+ Health Unit 2000*

20% ±1.7

400* 20 ±3.8

1600* 20±1.9

380* 17±0.9

76* 17±2.1

304* 17±1.0

Rural sector 1000 20 ±2.4

200 20±5.3

800 20 ±2.7

190 17±1.3

38 17±3.0

152 17±1.5

Urban sector 1000 20 ±2.4

200 20±5.3

800 20 ±2.7

190 17±1.3

38 17±3.0

152 17±1.5

Embargoed

Not for release: Preliminary analyses pending adjustment of variance estimates to account for complex survey design

Page 41: Complex Survey Samples Explaining the Miracle: Statistics and Analysis in Public Health APHEO Conference 2007, October 14-16, 2007 Susan Bondy, Department

41

Q & A