Proportions
proportions.RmdWe can estimate the proportions of the categories of any variable
using the function repprop(), as any other “rep” function
of ILSAstats, we need to specify the data
(df), the total weights (wt), the replicate
weights (repwt), and the method (method).
Besides these basic options, other arguments can be used:
-
x: a string with the name of the variable (or variables) to be used in the analysis. If multiple variables are specified, they will be treated as plausible values. -
group: a string containing the name of the variable that contains the groups of countries. If used all statistics will be estimated separately for each group. -
exclude: a string containing which groups should be excluded from aggregate estimations. -
aggregates: a string containing the aggregate statistics that should be estimated. Options include:"pooled"for also estimating all groups (without exclusions) as a single group; and"composite"for averaging all the estimations for each single group (without exclusions).
Weights and setup
For repprop(), first we need to create the replicate
weights. Using the included repdata data, and using the
"LANA" method:
RW <- repcreate(df = repdata,
wt = "wt",
jkzone = "jkzones",
jkrep = "jkrep",
method = "LANA")To make it easier to specify some arguments, it is advised that we
create also a "repsetup" object. We will create three
setups for this example: one without groups, one with groups and without
exclusions, and one with groups and exclusions (excluding group 2):
# No groups
STNG <- repsetup(repwt = RW, wt = "wt", df = repdata, method = "LANA")
# With groups
STGR <- repsetup(repwt = RW, wt = "wt", df = repdata, method = "LANA",
group = "GROUP")
# With groups and exclusions
STGE <- repsetup(repwt = RW, wt = "wt", df = repdata, method = "LANA",
group = "GROUP", exclude = "GR2")Single variable
For example, if we want to estimate the proportions of each category
of variable "GENDER", we can use either of the setups to
get the overall or group results:
# No groups
repprop(x = "GENDER", setup = STNG)## $`GENDER==0`
## N prop se prop_1 propdiff_1 propdiffse_1 tvalue_1
## 1 2483 0.49535 0.00773 0.50465 -0.00929 0.01546 -0.6012
##
## $`GENDER==1`
## N prop se prop_0 propdiff_0 propdiffse_0 tvalue_0
## 1 2517 0.50465 0.00773 0.49535 0.00929 0.01546 0.6012
# With groups
repprop(x = "GENDER", setup = STGR)## $`GENDER==0`
## group N prop se prop_1 propdiff_1 propdiffse_1 tvalue_1
## 1 Pooled 2483 0.49535 0.00773 0.50465 -0.00929 0.01546 -0.60120
## 2 Composite NA 0.49533 0.00774 0.50467 -0.00933 0.01548 -0.60308
## 3 GR1 819 0.48976 0.01298 0.51024 -0.02048 0.02596 -0.78907
## 4 GR2 846 0.50590 0.01248 0.49410 0.01180 0.02496 0.47286
## 5 GR3 818 0.49034 0.01465 0.50966 -0.01932 0.02931 -0.65928
##
## $`GENDER==1`
## group N prop se prop_0 propdiff_0 propdiffse_0 tvalue_0
## 1 Pooled 2517 0.50465 0.00773 0.49535 0.00929 0.01546 0.60120
## 2 Composite NA 0.50467 0.00774 0.49533 0.00933 0.01548 0.60308
## 3 GR1 848 0.51024 0.01298 0.48976 0.02048 0.02596 0.78907
## 4 GR2 820 0.49410 0.01248 0.50590 -0.01180 0.02496 -0.47286
## 5 GR3 849 0.50966 0.01465 0.49034 0.01932 0.02931 0.65928
# With groups and exclusions
repprop(x = "GENDER", setup = STGE)## $`GENDER==0`
## group N prop se prop_1 propdiff_1 propdiffse_1 tvalue_1
## 1 Pooled 1637 0.49005 0.01071 0.50995 -0.01990 0.02143 -0.92882
## 2 Composite NA 0.49005 0.00979 0.50995 -0.01990 0.01957 -1.01671
## 3 GR1 819 0.48976 0.01298 0.51024 -0.02048 0.02596 -0.78907
## 4 GR2 846 0.50590 0.01248 0.49410 0.01180 0.02496 0.47286
## 5 GR3 818 0.49034 0.01465 0.50966 -0.01932 0.02931 -0.65928
##
## $`GENDER==1`
## group N prop se prop_0 propdiff_0 propdiffse_0 tvalue_0
## 1 Pooled 1697 0.50995 0.01071 0.49005 0.01990 0.02143 0.92882
## 2 Composite NA 0.50995 0.00979 0.49005 0.01990 0.01957 1.01671
## 3 GR1 848 0.51024 0.01298 0.48976 0.02048 0.02596 0.78907
## 4 GR2 820 0.49410 0.01248 0.50590 -0.01180 0.02496 -0.47286
## 5 GR3 849 0.50966 0.01465 0.49034 0.01932 0.02931 0.65928We can notice that using no groups we would get the same results for the pooled estimates if we use groups and no exclusions. But, when we exclude group 2, the pooled and the composite estimate changes.
Proportion table
By default, repprop() will provide a list where each
element corresponds to the statistics of each category. Nevertheless, we
can summarize these results into a single data frame using
repprop.table():
p1 <- repprop(x = "GENDER", setup = STGR)
repprop.table(x = p1)## group category prop se
## 1 Pooled 0 0.49535 0.00773
## 2 Composite 0 0.49533 0.00774
## 3 GR1 0 0.48976 0.01298
## 4 GR2 0 0.50590 0.01248
## 5 GR3 0 0.49034 0.01465
## 6 Pooled 1 0.50465 0.00773
## 7 Composite 1 0.50467 0.00774
## 8 GR1 1 0.51024 0.01298
## 9 GR2 1 0.49410 0.01248
## 10 GR3 1 0.50966 0.01465By default, repprop.table() will provide a long table,
but we can also get wide tables were groups are separated by columns or
by rows, and combined or not the proportion estimates with the standard
errors:
# Groups by rows, separate SE
repprop.table(x = p1, type = "wide1")## $prop
## group prop.0 prop.1
## 1 Pooled 0.49535 0.50465
## 2 Composite 0.49533 0.50467
## 3 GR1 0.48976 0.51024
## 4 GR2 0.50590 0.49410
## 5 GR3 0.49034 0.50966
##
## $se
## group se.0 se.1
## 1 Pooled 0.00773 0.00773
## 2 Composite 0.00774 0.00774
## 3 GR1 0.01298 0.01298
## 4 GR2 0.01248 0.01248
## 5 GR3 0.01465 0.01465
# Groups by rows, non-separate SE
repprop.table(x = p1, type = "wide1", separateSE = FALSE)## group prop.0 se.0 prop.1 se.1
## 1 Pooled 0.49535 0.00773 0.50465 0.00773
## 2 Composite 0.49533 0.00774 0.50467 0.00774
## 3 GR1 0.48976 0.01298 0.51024 0.01298
## 4 GR2 0.50590 0.01248 0.49410 0.01248
## 5 GR3 0.49034 0.01465 0.50966 0.01465
# Groups by columns, separate SE
repprop.table(x = p1, type = "wide2")## $prop
## category Pooled Composite GR1 GR2 GR3
## 1 0 0.4953532 0.4953333 0.4897593 0.5059013 0.4903392
## 3 1 0.5046468 0.5046667 0.5102407 0.4940987 0.5096608
##
## $se
## category Pooled Composite GR1 GR2 GR3
## 2 0 0.007729275 0.007738154 0.01297812 0.01248006 0.01465359
## 4 1 0.007729275 0.007738154 0.01297812 0.01248006 0.01465359
# Groups by columns, non-separate SE
repprop.table(x = p1, type = "wide2", separateSE = FALSE)## category statistic Pooled Composite GR1 GR2 GR3
## 1 0 prop 0.495353153 0.495333268 0.48975930 0.50590131 0.49033920
## 2 0 se 0.007729275 0.007738154 0.01297812 0.01248006 0.01465359
## 3 1 prop 0.504646847 0.504666732 0.51024070 0.49409869 0.50966080
## 4 1 se 0.007729275 0.007738154 0.01297812 0.01248006 0.01465359Plausible values
When treating with plausible values, we need to specify the names of all plausible values of a construct, so all estimates will be combined. For example, for estimating the proportions of each proficiency level in math we would use:
mean achievement in math for this sample we would use:
# No groups
repprop(x = paste0("CatMath",1:5), setup = STNG)|>
repprop.table(type = "wide2")## More than one variable provided. 'x' treated as PVs.## $prop
## category Pooled
## 1 1 0.1604335
## 3 2 0.3388836
## 5 3 0.3343512
## 7 4 0.1663317
##
## $se
## category Pooled
## 2 1 0.008401597
## 4 2 0.009643329
## 6 3 0.010323985
## 8 4 0.006739155
# With groups
repprop(x = paste0("CatMath",1:5), setup = STGR)|>
repprop.table(type = "wide2")## More than one variable provided. 'x' treated as PVs.## $prop
## category Pooled Composite GR1 GR2 GR3
## 1 1 0.1604335 0.1605950 0.31764327 0.1230013 0.04114025
## 3 2 0.3388836 0.3388807 0.42791017 0.3791503 0.20958161
## 5 3 0.3343512 0.3342311 0.21615471 0.3616510 0.42488755
## 7 4 0.1663317 0.1662933 0.03829185 0.1361973 0.32439059
##
## $se
## category Pooled Composite GR1 GR2 GR3
## 2 1 0.008401597 0.008045957 0.020893302 0.01068229 0.00565645
## 4 2 0.009643329 0.009302842 0.016750950 0.01374063 0.01759223
## 6 3 0.010323985 0.010593748 0.015851628 0.02033506 0.01858114
## 8 4 0.006739155 0.008037791 0.006531979 0.01193100 0.01991078Aggregates
When using groups we can always omit the pooled and composite calculations if we need to, by default both estimates will be calculated.
# Default
repprop(x = paste0("CatMath",1:5), setup = STGR)|>
repprop.table(type = "wide2",separateSE = FALSE)## More than one variable provided. 'x' treated as PVs.## category statistic Pooled Composite GR1 GR2 GR3
## 1 1 prop 0.160433532 0.160594953 0.317643271 0.12300134 0.04114025
## 2 1 se 0.008401597 0.008045957 0.020893302 0.01068229 0.00565645
## 3 2 prop 0.338883595 0.338880700 0.427910174 0.37915032 0.20958161
## 4 2 se 0.009643329 0.009302842 0.016750950 0.01374063 0.01759223
## 5 3 prop 0.334351157 0.334231088 0.216154706 0.36165101 0.42488755
## 6 3 se 0.010323985 0.010593748 0.015851628 0.02033506 0.01858114
## 7 4 prop 0.166331717 0.166293259 0.038291849 0.13619733 0.32439059
## 8 4 se 0.006739155 0.008037791 0.006531979 0.01193100 0.01991078
# Only pooled
repprop(x = paste0("CatMath",1:5), setup = STGR, aggregates = "pooled")|>
repprop.table(type = "wide2",separateSE = FALSE)## More than one variable provided. 'x' treated as PVs.## category statistic Pooled GR1 GR2 GR3
## 1 1 prop 0.160433532 0.317643271 0.12300134 0.04114025
## 2 1 se 0.008401597 0.020893302 0.01068229 0.00565645
## 3 2 prop 0.338883595 0.427910174 0.37915032 0.20958161
## 4 2 se 0.009643329 0.016750950 0.01374063 0.01759223
## 5 3 prop 0.334351157 0.216154706 0.36165101 0.42488755
## 6 3 se 0.010323985 0.015851628 0.02033506 0.01858114
## 7 4 prop 0.166331717 0.038291849 0.13619733 0.32439059
## 8 4 se 0.006739155 0.006531979 0.01193100 0.01991078
# Only composite
repprop(x = paste0("CatMath",1:5), setup = STGR, aggregates = "composite")|>
repprop.table(type = "wide2",separateSE = FALSE)## More than one variable provided. 'x' treated as PVs.## category statistic Composite GR1 GR2 GR3
## 1 1 prop 0.160594953 0.317643271 0.12300134 0.04114025
## 2 1 se 0.008045957 0.020893302 0.01068229 0.00565645
## 3 2 prop 0.338880700 0.427910174 0.37915032 0.20958161
## 4 2 se 0.009302842 0.016750950 0.01374063 0.01759223
## 5 3 prop 0.334231088 0.216154706 0.36165101 0.42488755
## 6 3 se 0.010593748 0.015851628 0.02033506 0.01858114
## 7 4 prop 0.166293259 0.038291849 0.13619733 0.32439059
## 8 4 se 0.008037791 0.006531979 0.01193100 0.01991078
# No aggregates
repprop(x = paste0("CatMath",1:5), setup = STGR, aggregates = NULL)|>
repprop.table(type = "wide2",separateSE = FALSE)## More than one variable provided. 'x' treated as PVs.## category statistic GR1 GR2 GR3
## 1 1 prop 0.317643271 0.12300134 0.04114025
## 2 1 se 0.020893302 0.01068229 0.00565645
## 3 2 prop 0.427910174 0.37915032 0.20958161
## 4 2 se 0.016750950 0.01374063 0.01759223
## 5 3 prop 0.216154706 0.36165101 0.42488755
## 6 3 se 0.015851628 0.02033506 0.01858114
## 7 4 prop 0.038291849 0.13619733 0.32439059
## 8 4 se 0.006531979 0.01193100 0.01991078