Skip to contents

Estimate all standard RAM-OP indicators

Usage

estimate_op_all(
  x,
  w,
  indicators = c("demo", "anthro", "food", "hunger", "adl", "disability", "mental",
    "dementia", "health", "oedema", "screening", "income", "wash", "visual", "misc"),
  replicates = 399
)

Arguments

x

Indicators dataset produced by create_op_all with primary sampling unit (PSU) in column named PSU

w

A data frame with primary sampling unit (PSU) in column named psu and survey weight (i.e. PSU population) in column named pop

indicators

A character vector of indicator set names to estimate. Indicator set names are demo, anthro, food, hunger, disability, adl, mental, dementia, health, income, wash, visual, and misc. Default is all indicator sets.

replicates

Number of bootstrap replicates. Default is 399.

Value

Tibble of boot estimates for all specified standard RAM-OP indicators

Examples

estimate_op_all(x = create_op_all(testSVY),
                w = testPSU,
                replicates = 9)
#> # A tibble: 139 × 13
#>    INDICATOR GROUP       LABEL TYPE  EST.ALL LCL.ALL UCL.ALL EST.MALES LCL.MALES
#>    <fct>     <fct>       <fct> <fct>   <dbl>   <dbl>   <dbl>     <dbl>     <dbl>
#>  1 resp1     Survey      Resp… Prop…   87.0   81.5     91.7      84        75.8 
#>  2 resp2     Survey      Resp… Prop…    7.81   5.83    10.9       7.89      2.48
#>  3 resp3     Survey      Resp… Prop…    3.65   0.729    6.67      6.33      1.55
#>  4 resp4     Survey      Resp… Prop…    1.04   0        2.08      1.27      0   
#>  5 age       Demography… Mean… Mean    70.2   69.0     71.7      70.2      69.1 
#>  6 ageGrp1   Demography… Self… Prop…    0      0        0         0         0   
#>  7 ageGrp2   Demography… Self… Prop…   54.2   48.0     62.2      50        39.6 
#>  8 ageGrp3   Demography… Self… Prop…   22.4   19.5     33.3      27.1      17.2 
#>  9 ageGrp4   Demography… Self… Prop…   18.8   11.5     26.1      18.4       6.86
#> 10 ageGrp5   Demography… Self… Prop…    2.08   0.729    4.06      5.33      1.46
#> # ℹ 129 more rows
#> # ℹ 4 more variables: UCL.MALES <dbl>, EST.FEMALES <dbl>, LCL.FEMALES <dbl>,
#> #   UCL.FEMALES <dbl>