Lecture 1-3 Sampling & weights

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/35

encourage image

There's no tags or description

Looks like no tags are added yet.

36 Terms

1
New cards

Target population

The full group the survey wants to learn about

2
New cards

Sampling Frame

The list/ material from which the sample is actually drawn

3
New cards

Sampled population

The units that had a chance of selection from the sampling frame

  • Can differ from the target population (coverage error)


4
New cards

Probability Sample

Every unit has a known, non-zero chance of selection, so design weights and population inference are possible

5
New cards

Non-probability sample

Selection probabilities are unknown (web polls e.g.), so no design weights and no poulation inference.

6
New cards

Simple Random Sample (SRS)

Every possible Sample of size n is equally likely

7
New cards

Simple Random Sample With Replacement (SRSWR)

A unit can be drawn more than once:

  • Pr(selection) = 1 - (1-1/N)^n


8
New cards

Simple Random Sample Without Replacement

A unit can be drawn once only

  • Pr(selection) = n/N


9
New cards

Stratified Sampling

Divide the population into non-overlapping strata and sample independently

10
New cards

Proportionate stratification

Sampling fraction fh = nh/Nh is the same in every stratum

11
New cards

Disproportionate Stratificaiton

Sampling fractions differ across strata (e.g. oversampling a small group or large firms), needs weights

12
New cards

Cluster Sampling

Sample groups (clusters) of units, not individuals; only some clusters are selected

13
New cards

One-stage cluster sample

All units in each selected cluster are surveyed

14
New cards

Two-stage cluster sample

Select clusters (PSU), then sample units within each selected cluster

15
New cards

Multi-stage sample

Further random selection at more stages (e.g. EA, then household, then person)

16
New cards

Primary Sampling Unit (PSU)

The first-stage unit sampled (e.g. in household survey, the EA)

17
New cards

Enumeration Area (EA)

Small census geographic area used as the cluster/ PSU

18
New cards

Systematic Sampling

Pick a random start, then take very k-th unit form a list (k = N/n)

19
New cards

Implicit stratification

Sort the list by variable, then sample systematically so the sample spreads across that variable

20
New cards

Probability Proportional to Size (PPS)

Selection probability of a cluster is proportional to its size Nj

21
New cards

Equal Probability of Selection Method

Every unit in the population has the same selection probability, so weights ar all equal

22
New cards

Listing

Enumerating all dwellings/ household in a selected cluster so a second-stage sample can be drawn

23
New cards

Stratified two-stage design (SA hh surveys)

Strata (province x metro/non-metro x geotype), PPS selection of PSUs, then systematic selection of dwellings within PSUs

24
New cards

Design Weight

1/πi the inverse of the units probability of selection,

  • The number of population units that household represents


25
New cards

Non-response weight

Design weight x 1/(response rate of units group)

  • Respondents are scaled up to represent non-respondents


26
New cards

Unit non-response

A sampled unit gives no data at all

27
New cards

Item non-response

A responding unit skips particular questions

28
New cards

Missing Completely at Random (MCAR)

Missingness is unrelated to any observed or unobserved variable

29
New cards

Missing at Random (MAR)

Missingness depends on observed variables only and can be ignored after conditioning on them

30
New cards

Not Missing at Random (NMAR)

Missingness depends on the unobserved value itself, even after conditioning on observed variables

31
New cards

pweight

Stata probability weight, correct point estimate and with svyset, correct SEs

32
New cards

aweight

Analytic weight, rescaled to sum to one, correct point estimate but SEs not design-correct. Use only when a command wont take pweight

33
New cards

fweight

Frequency weight, each observation is replicated that many times (integer counts0

34
New cards

iweight

Imprtance weight, raw weight with no statisical interpretaitons, some commands need it like povdeco

35
New cards

svyset

Tells Stata the design (weights, PSUs, strata)l so svy: commands give design correct SEs

36
New cards

Nationally Representative

Probability sample from a frame covering the whole target population properly weighted, with Ses small enough to be useful