sdd-disability-statistics-toolkit/src/disability-analysis-examples.R at main · PacificCommunity/sdd-disability-statistics-toolkit · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
# Set-up
library(tidyverse)
library(survey)

# Note: this is completely synthetic, outputs may not make sense
data <- read.csv('raw_data/example-disability-data.csv')

# Coding indicators

data <- data |>
  mutate(primary_education = NA)

data <- data |>
  mutate(primary_education = ifelse(age >= 15, 0, primary_education))

data <- data |>
  mutate(primary_education = ifelse(primary_education %in% 0
                                    & grade_completed %in% 16:45,
                                    1, primary_education))
data <- data |>
  mutate(primary_education = ifelse(age >= 15
                                    & is.na(grade_completed),
                                    0, primary_education))

# Coding categorical variables

data <- data |>
  mutate(age_groups = case_when(
    age >= 5 & age < 18 ~ "5 to 17",
    age >= 18 & age < 50 ~ "18 to 49",
    age >= 50 ~ "50+"
  ),
  age_groups = factor(age_groups,
                      levels = c("5 to 17","18 to 49","50+")))

data <- data |>
  mutate(disability = NA,
         disability = ifelse(!is.na(seeing) & !is.na(hearing)
                             & !is.na(mobility) & !is.na(memory)
                             & !is.na(selfcare) & !is.na(communication)
                             & age >= 5, 0, disability),
         disability = ifelse(disability %in% 0
                             & (seeing %in% 3:4
                                | hearing %in% 3:4
                                | mobility %in% 3:4
                                | memory %in% 3:4
                                | selfcare %in% 3:4
                                | communication %in% 3:4),
                             1, disability))

# Calculating absolute numbers

data |> count(disability)

data |> count(disability, age_groups)

data |> count(disability, age_groups, sex)

data |> count(disability, sex, wt = sweight)

# Estimating measures of prevalance

data |>
  filter(!is.na(primary_education)) |>
  count(primary_education) |>
  mutate(prop = n/sum(n))

data |>
  filter(!is.na(disability)) |>
  count(disability) |>
  mutate(prop = n/sum(n))

# Utilising the survey packag

design <- svydesign(ids = ~cluster, weights = ~sweight,
                    strata = ~strata, data = data)
svyciprop(~I(disability == 1), design, method = "logit", na.rm = TRUE)

# Stratifying measures of prevalance

data |>
  filter(!is.na(disability), !is.na(age_groups)) |>
  group_by(age_groups) |>
  count(disability) |>
  mutate(prop = n/sum(n))


data |>
  filter(!is.na(disability),
         !is.na(sex),
         !is.na(primary_education)) |>
  group_by(disability, sex) |>
  count(primary_education) |>
  mutate(prop = n/sum(n))

design <- svydesign(ids = ~cluster, weights = ~sweight,
                    strata = ~strata, data = data)

svyby(~I(primary_education == 1), ~sex + disability, design,
      svyciprop, method = "logit", vartype = "ci", na.rm = TRUE)