Two small, partially overlapping datasets, built to mimick non-independent PSU samples selected with a one- or multi-stage, stratified, cluster sampling design (but identifiers of SSUs etc. are not reported). Allow to run R code contained in the ‘Examples’ section of ReGenesees function svyDelta.

data(Delta.clus)

Format

Two data frames, sclus1 and sclus2, with 6 PSUs each (and 20 and 22 final units, respectively), and the following 5 variables.

  • For both samples sclus1 and sclus2:

id

Identifier of sample PSUs, numeric

strata

Stratification variable, a factor with 2 levels: A, and B

w

Sampling weights of final units, numeric

y

A numeric variable

x

A numeric variable, correlated with y

Details

The two samples, sclus1 and sclus2, have 3 PSUs in common, resulting in an overlap rate of 3 / 6 = 0.5 at PSU-level. One could think of them as, e.g., two consecutive waves of a rotating panel with a 50% overlap at PSU-level.

Common PSUs are unambigously identified by variable id.

The stratification is static: (1) sclus1 and sclus2 use the same strata (i.e. levels A, and B), and (2) no common PSUs changed stratum from sclus1 to sclus2.

The ‘Examples’ section of svyDelta will illustrate the effect of dynamic stratification by injecting new strata and stratum-changer units in the samples.

See also

svyDelta for calculating estimates and sampling errors of Measures of Change from two not necessarily independent samples, and Delta.el for 2 artificial overlapping samples of elementary units.

Examples

data(Delta.clus) # Have a look: sclus1
#> id strata w y x #> 1 1 A 18.55982 0.5 0.9299534 #> 2 1 A 21.06586 1.0 2.0829387 #> 3 1 A 21.93166 1.5 3.0842540 #> 4 2 A 19.98340 2.0 0.8722675 #> 5 2 A 18.51504 2.5 3.3248515 #> 6 3 A 20.28786 3.0 -0.3339274 #> 7 3 A 19.47544 3.5 3.8162962 #> 8 3 A 20.08048 4.0 2.2946235 #> 9 3 A 19.35298 4.5 3.5662329 #> 10 4 B 21.64282 5.0 4.5246152 #> 11 4 B 30.33911 5.5 3.3212096 #> 12 4 B 29.99442 6.0 3.8279945 #> 13 4 B 31.15830 6.5 3.9486667 #> 14 5 B 30.75910 7.0 5.1722682 #> 15 5 B 29.87766 7.5 4.3871676 #> 16 5 B 29.97200 8.0 6.2267104 #> 17 5 B 29.66927 8.5 4.4201194 #> 18 5 B 30.54008 9.0 5.7321373 #> 19 6 B 30.21610 9.5 3.6582305 #> 20 6 B 30.05322 10.0 5.6449544
sclus2
#> id strata w y x #> 1 7 A 18.07945 2.512918 2.078354 #> 2 7 A 19.16909 3.308235 2.686179 #> 3 7 A 19.98976 5.003790 3.743181 #> 4 7 A 19.30035 2.977190 2.923050 #> 5 8 A 20.28916 4.510136 3.046503 #> 6 8 A 20.57337 3.165896 3.431041 #> 7 8 A 20.46191 3.504128 3.298700 #> 8 3 A 20.28786 4.540171 4.835238 #> 9 3 A 19.47544 4.985899 2.876468 #> 10 3 A 20.08048 6.370713 3.967471 #> 11 3 A 19.35298 6.681330 4.072293 #> 12 4 B 21.64282 7.067339 4.694993 #> 13 4 B 30.33911 6.551161 2.886048 #> 14 4 B 29.99442 8.740280 5.930593 #> 15 4 B 31.15830 8.118962 5.003309 #> 16 11 B 29.54043 8.900935 4.997548 #> 17 11 B 31.38035 9.446146 6.679908 #> 18 11 B 31.18770 9.596294 7.131485 #> 19 11 B 29.42232 9.444555 5.482839 #> 20 11 B 29.59098 10.408366 6.806784 #> 21 6 B 30.21610 12.512506 7.079508 #> 22 6 B 30.05322 11.124746 7.090510
# Have a look at the overlap subsample of 3 PSUs (36 final units): sc <- merge(sclus1, sclus2, by = "id", suffixes = c("1", "2")) sc
#> id strata1 w1 y1 x1 strata2 w2 y2 x2 #> 1 3 A 19.47544 3.5 3.8162962 A 19.47544 4.985899 2.876468 #> 2 3 A 19.47544 3.5 3.8162962 A 20.08048 6.370713 3.967471 #> 3 3 A 19.47544 3.5 3.8162962 A 19.35298 6.681330 4.072293 #> 4 3 A 19.47544 3.5 3.8162962 A 20.28786 4.540171 4.835238 #> 5 3 A 20.08048 4.0 2.2946235 A 19.47544 4.985899 2.876468 #> 6 3 A 20.08048 4.0 2.2946235 A 20.08048 6.370713 3.967471 #> 7 3 A 20.08048 4.0 2.2946235 A 19.35298 6.681330 4.072293 #> 8 3 A 20.08048 4.0 2.2946235 A 20.28786 4.540171 4.835238 #> 9 3 A 19.35298 4.5 3.5662329 A 19.47544 4.985899 2.876468 #> 10 3 A 19.35298 4.5 3.5662329 A 20.08048 6.370713 3.967471 #> 11 3 A 19.35298 4.5 3.5662329 A 19.35298 6.681330 4.072293 #> 12 3 A 19.35298 4.5 3.5662329 A 20.28786 4.540171 4.835238 #> 13 3 A 20.28786 3.0 -0.3339274 A 19.47544 4.985899 2.876468 #> 14 3 A 20.28786 3.0 -0.3339274 A 20.08048 6.370713 3.967471 #> 15 3 A 20.28786 3.0 -0.3339274 A 19.35298 6.681330 4.072293 #> 16 3 A 20.28786 3.0 -0.3339274 A 20.28786 4.540171 4.835238 #> 17 4 B 30.33911 5.5 3.3212096 B 30.33911 6.551161 2.886048 #> 18 4 B 30.33911 5.5 3.3212096 B 29.99442 8.740280 5.930593 #> 19 4 B 30.33911 5.5 3.3212096 B 31.15830 8.118962 5.003309 #> 20 4 B 30.33911 5.5 3.3212096 B 21.64282 7.067339 4.694993 #> 21 4 B 29.99442 6.0 3.8279945 B 30.33911 6.551161 2.886048 #> 22 4 B 29.99442 6.0 3.8279945 B 29.99442 8.740280 5.930593 #> 23 4 B 29.99442 6.0 3.8279945 B 31.15830 8.118962 5.003309 #> 24 4 B 29.99442 6.0 3.8279945 B 21.64282 7.067339 4.694993 #> 25 4 B 31.15830 6.5 3.9486667 B 30.33911 6.551161 2.886048 #> 26 4 B 31.15830 6.5 3.9486667 B 29.99442 8.740280 5.930593 #> 27 4 B 31.15830 6.5 3.9486667 B 31.15830 8.118962 5.003309 #> 28 4 B 31.15830 6.5 3.9486667 B 21.64282 7.067339 4.694993 #> 29 4 B 21.64282 5.0 4.5246152 B 30.33911 6.551161 2.886048 #> 30 4 B 21.64282 5.0 4.5246152 B 29.99442 8.740280 5.930593 #> 31 4 B 21.64282 5.0 4.5246152 B 31.15830 8.118962 5.003309 #> 32 4 B 21.64282 5.0 4.5246152 B 21.64282 7.067339 4.694993 #> 33 6 B 30.21610 9.5 3.6582305 B 30.21610 12.512506 7.079508 #> 34 6 B 30.21610 9.5 3.6582305 B 30.05322 11.124746 7.090510 #> 35 6 B 30.05322 10.0 5.6449544 B 30.21610 12.512506 7.079508 #> 36 6 B 30.05322 10.0 5.6449544 B 30.05322 11.124746 7.090510
# Have a look at the full rotation structure (50% PSUs overlap in each stratum): s <- merge(sclus1, sclus2, by = "id", all = TRUE, suffixes = c("1", "2")) s <- s[order(s$strata1, s$strata2), ] s
#> id strata1 w1 y1 x1 strata2 w2 y2 x2 #> 6 3 A 19.47544 3.5 3.8162962 A 19.47544 4.985899 2.876468 #> 7 3 A 19.47544 3.5 3.8162962 A 20.08048 6.370713 3.967471 #> 8 3 A 19.47544 3.5 3.8162962 A 19.35298 6.681330 4.072293 #> 9 3 A 19.47544 3.5 3.8162962 A 20.28786 4.540171 4.835238 #> 10 3 A 20.08048 4.0 2.2946235 A 19.47544 4.985899 2.876468 #> 11 3 A 20.08048 4.0 2.2946235 A 20.08048 6.370713 3.967471 #> 12 3 A 20.08048 4.0 2.2946235 A 19.35298 6.681330 4.072293 #> 13 3 A 20.08048 4.0 2.2946235 A 20.28786 4.540171 4.835238 #> 14 3 A 19.35298 4.5 3.5662329 A 19.47544 4.985899 2.876468 #> 15 3 A 19.35298 4.5 3.5662329 A 20.08048 6.370713 3.967471 #> 16 3 A 19.35298 4.5 3.5662329 A 19.35298 6.681330 4.072293 #> 17 3 A 19.35298 4.5 3.5662329 A 20.28786 4.540171 4.835238 #> 18 3 A 20.28786 3.0 -0.3339274 A 19.47544 4.985899 2.876468 #> 19 3 A 20.28786 3.0 -0.3339274 A 20.08048 6.370713 3.967471 #> 20 3 A 20.28786 3.0 -0.3339274 A 19.35298 6.681330 4.072293 #> 21 3 A 20.28786 3.0 -0.3339274 A 20.28786 4.540171 4.835238 #> 1 1 A 18.55982 0.5 0.9299534 <NA> NA NA NA #> 2 1 A 21.06586 1.0 2.0829387 <NA> NA NA NA #> 3 1 A 21.93166 1.5 3.0842540 <NA> NA NA NA #> 4 2 A 19.98340 2.0 0.8722675 <NA> NA NA NA #> 5 2 A 18.51504 2.5 3.3248515 <NA> NA NA NA #> 22 4 B 30.33911 5.5 3.3212096 B 30.33911 6.551161 2.886048 #> 23 4 B 30.33911 5.5 3.3212096 B 29.99442 8.740280 5.930593 #> 24 4 B 30.33911 5.5 3.3212096 B 31.15830 8.118962 5.003309 #> 25 4 B 30.33911 5.5 3.3212096 B 21.64282 7.067339 4.694993 #> 26 4 B 29.99442 6.0 3.8279945 B 30.33911 6.551161 2.886048 #> 27 4 B 29.99442 6.0 3.8279945 B 29.99442 8.740280 5.930593 #> 28 4 B 29.99442 6.0 3.8279945 B 31.15830 8.118962 5.003309 #> 29 4 B 29.99442 6.0 3.8279945 B 21.64282 7.067339 4.694993 #> 30 4 B 31.15830 6.5 3.9486667 B 30.33911 6.551161 2.886048 #> 31 4 B 31.15830 6.5 3.9486667 B 29.99442 8.740280 5.930593 #> 32 4 B 31.15830 6.5 3.9486667 B 31.15830 8.118962 5.003309 #> 33 4 B 31.15830 6.5 3.9486667 B 21.64282 7.067339 4.694993 #> 34 4 B 21.64282 5.0 4.5246152 B 30.33911 6.551161 2.886048 #> 35 4 B 21.64282 5.0 4.5246152 B 29.99442 8.740280 5.930593 #> 36 4 B 21.64282 5.0 4.5246152 B 31.15830 8.118962 5.003309 #> 37 4 B 21.64282 5.0 4.5246152 B 21.64282 7.067339 4.694993 #> 43 6 B 30.21610 9.5 3.6582305 B 30.21610 12.512506 7.079508 #> 44 6 B 30.21610 9.5 3.6582305 B 30.05322 11.124746 7.090510 #> 45 6 B 30.05322 10.0 5.6449544 B 30.21610 12.512506 7.079508 #> 46 6 B 30.05322 10.0 5.6449544 B 30.05322 11.124746 7.090510 #> 38 5 B 30.75910 7.0 5.1722682 <NA> NA NA NA #> 39 5 B 29.87766 7.5 4.3871676 <NA> NA NA NA #> 40 5 B 29.97200 8.0 6.2267104 <NA> NA NA NA #> 41 5 B 29.66927 8.5 4.4201194 <NA> NA NA NA #> 42 5 B 30.54008 9.0 5.7321373 <NA> NA NA NA #> 47 7 <NA> NA NA NA A 18.07945 2.512918 2.078354 #> 48 7 <NA> NA NA NA A 19.16909 3.308235 2.686179 #> 49 7 <NA> NA NA NA A 19.98976 5.003790 3.743181 #> 50 7 <NA> NA NA NA A 19.30035 2.977190 2.923050 #> 51 8 <NA> NA NA NA A 20.28916 4.510136 3.046503 #> 52 8 <NA> NA NA NA A 20.57337 3.165896 3.431041 #> 53 8 <NA> NA NA NA A 20.46191 3.504128 3.298700 #> 54 11 <NA> NA NA NA B 29.54043 8.900935 4.997548 #> 55 11 <NA> NA NA NA B 31.38035 9.446146 6.679908 #> 56 11 <NA> NA NA NA B 31.18770 9.596294 7.131485 #> 57 11 <NA> NA NA NA B 29.42232 9.444555 5.482839 #> 58 11 <NA> NA NA NA B 29.59098 10.408366 6.806784
# As anticipated, strata are static: with(s, table(strata1, strata2, useNA = "ifany"))
#> strata2 #> strata1 A B <NA> #> A 16 0 5 #> B 0 20 5 #> <NA> 7 5 0