Primeiramente consumimos os dados a serem utilizados para a análise

data = read_csv("speed-dating/speed-dating2.csv")
## Parsed with column specification:
## cols(
##   .default = col_double(),
##   field = col_character(),
##   from = col_character(),
##   career = col_character(),
##   attr3_s = col_logical(),
##   sinc3_s = col_logical(),
##   intel3_s = col_logical(),
##   fun3_s = col_logical(),
##   amb3_s = col_logical(),
##   dec = col_character()
## )
## See spec(...) for full column specifications.
## Warning: 10220 parsing failures.
##  row      col           expected actual                             file
## 1847 attr3_s  1/0/T/F/TRUE/FALSE  8.00  'speed-dating/speed-dating2.csv'
## 1847 sinc3_s  1/0/T/F/TRUE/FALSE  10.00 'speed-dating/speed-dating2.csv'
## 1847 intel3_s 1/0/T/F/TRUE/FALSE  9.00  'speed-dating/speed-dating2.csv'
## 1847 fun3_s   1/0/T/F/TRUE/FALSE  10    'speed-dating/speed-dating2.csv'
## 1847 amb3_s   1/0/T/F/TRUE/FALSE  10    'speed-dating/speed-dating2.csv'
## .... ........ .................. ...... ................................
## See problems(...) for more details.
data = data %>% mutate(match = ifelse(dec == "yes", 1, 0))

EDA

skimr::skim(data)
variable type stat level value formatted
iid numeric missing .all 0.0000000 0
iid numeric complete .all 4918.0000000 4918
iid numeric n .all 4918.0000000 4918
iid numeric mean .all 274.6748678 274.67
iid numeric sd .all 183.9075961 183.91
iid numeric p0 .all 1.0000000 1
iid numeric p25 .all 88.0000000 88
iid numeric p50 .all 273.0000000 273
iid numeric p75 .all 431.0000000 431
iid numeric p100 .all 552.0000000 552
iid numeric hist .all NA ▇▇▁▆▃▆▃▇
gender numeric missing .all 0.0000000 0
gender numeric complete .all 4918.0000000 4918
gender numeric n .all 4918.0000000 4918
gender numeric mean .all 0.5010167 0.5
gender numeric sd .all 0.5000498 0.5
gender numeric p0 .all 0.0000000 0
gender numeric p25 .all 0.0000000 0
gender numeric p50 .all 1.0000000 1
gender numeric p75 .all 1.0000000 1
gender numeric p100 .all 1.0000000 1
gender numeric hist .all NA ▇▁▁▁▁▁▁▇
order numeric missing .all 0.0000000 0
order numeric complete .all 4918.0000000 4918
order numeric n .all 4918.0000000 4918
order numeric mean .all 9.2602684 9.26
order numeric sd .all 5.6694748 5.67
order numeric p0 .all 1.0000000 1
order numeric p25 .all 4.0000000 4
order numeric p50 .all 9.0000000 9
order numeric p75 .all 14.0000000 14
order numeric p100 .all 22.0000000 22
order numeric hist .all NA ▇▇▅▇▅▃▅▂
pid numeric missing .all 10.0000000 10
pid numeric complete .all 4908.0000000 4908
pid numeric n .all 4918.0000000 4918
pid numeric mean .all 274.9767726 274.98
pid numeric sd .all 183.9730364 183.97
pid numeric p0 .all 1.0000000 1
pid numeric p25 .all 88.0000000 88
pid numeric p50 .all 273.0000000 273
pid numeric p75 .all 431.0000000 431
pid numeric p100 .all 552.0000000 552
pid numeric hist .all NA ▇▇▁▆▃▆▃▇
int_corr numeric missing .all 72.0000000 72
int_corr numeric complete .all 4846.0000000 4846
int_corr numeric n .all 4918.0000000 4918
int_corr numeric mean .all 0.1905530 0.19
int_corr numeric sd .all 0.3085869 0.31
int_corr numeric p0 .all -0.7300000 -0.73
int_corr numeric p25 .all -0.0300000 -0.03
int_corr numeric p50 .all 0.2100000 0.21
int_corr numeric p75 .all 0.4300000 0.43
int_corr numeric p100 .all 0.9000000 0.9
int_corr numeric hist .all NA ▁▂▅▆▇▇▅▁
samerace numeric missing .all 0.0000000 0
samerace numeric complete .all 4918.0000000 4918
samerace numeric n .all 4918.0000000 4918
samerace numeric mean .all 0.4062627 0.41
samerace numeric sd .all 0.4911847 0.49
samerace numeric p0 .all 0.0000000 0
samerace numeric p25 .all 0.0000000 0
samerace numeric p50 .all 0.0000000 0
samerace numeric p75 .all 1.0000000 1
samerace numeric p100 .all 1.0000000 1
samerace numeric hist .all NA ▇▁▁▁▁▁▁▆
age_o numeric missing .all 61.0000000 61
age_o numeric complete .all 4857.0000000 4857
age_o numeric n .all 4918.0000000 4918
age_o numeric mean .all 25.7906115 25.79
age_o numeric sd .all 3.3479583 3.35
age_o numeric p0 .all 18.0000000 18
age_o numeric p25 .all 23.0000000 23
age_o numeric p50 .all 25.0000000 25
age_o numeric p75 .all 28.0000000 28
age_o numeric p100 .all 39.0000000 39
age_o numeric hist .all NA ▁▇▆▇▃▁▁▁
age numeric missing .all 52.0000000 52
age numeric complete .all 4866.0000000 4866
age numeric n .all 4918.0000000 4918
age numeric mean .all 25.7813399 25.78
age numeric sd .all 3.3523272 3.35
age numeric p0 .all 18.0000000 18
age numeric p25 .all 23.0000000 23
age numeric p50 .all 25.0000000 25
age numeric p75 .all 28.0000000 28
age numeric p100 .all 39.0000000 39
age numeric hist .all NA ▁▇▆▇▃▁▁▁
field character missing .all 20.0000000 20
field character complete .all 4898.0000000 4898
field character n .all 4918.0000000 4918
field character min .all 3.0000000 3
field character max .all 51.0000000 51
field character empty .all 0.0000000 0
field character n_unique .all 148.0000000 148
race numeric missing .all 20.0000000 20
race numeric complete .all 4898.0000000 4898
race numeric n .all 4918.0000000 4918
race numeric mean .all 2.7311147 2.73
race numeric sd .all 1.2196805 1.22
race numeric p0 .all 1.0000000 1
race numeric p25 .all 2.0000000 2
race numeric p50 .all 2.0000000 2
race numeric p75 .all 4.0000000 4
race numeric p100 .all 6.0000000 6
race numeric hist .all NA ▁▇▁▁▃▁▁▁
from character missing .all 36.0000000 36
from character complete .all 4882.0000000 4882
from character n .all 4918.0000000 4918
from character min .all 2.0000000 2
from character max .all 58.0000000 58
from character empty .all 0.0000000 0
from character n_unique .all 172.0000000 172
career character missing .all 46.0000000 46
career character complete .all 4872.0000000 4872
career character n .all 4918.0000000 4918
career character min .all 2.0000000 2
career character max .all 77.0000000 77
career character empty .all 0.0000000 0
career character n_unique .all 218.0000000 218
sports numeric missing .all 36.0000000 36
sports numeric complete .all 4882.0000000 4882
sports numeric n .all 4918.0000000 4918
sports numeric mean .all 6.3961491 6.4
sports numeric sd .all 2.5661440 2.57
sports numeric p0 .all 1.0000000 1
sports numeric p25 .all 5.0000000 5
sports numeric p50 .all 7.0000000 7
sports numeric p75 .all 8.0000000 8
sports numeric p100 .all 10.0000000 10
sports numeric hist .all NA ▃▂▃▃▃▅▅▇
tvsports numeric missing .all 36.0000000 36
tvsports numeric complete .all 4882.0000000 4882
tvsports numeric n .all 4918.0000000 4918
tvsports numeric mean .all 4.5266284 4.53
tvsports numeric sd .all 2.8175787 2.82
tvsports numeric p0 .all 1.0000000 1
tvsports numeric p25 .all 2.0000000 2
tvsports numeric p50 .all 4.0000000 4
tvsports numeric p75 .all 7.0000000 7
tvsports numeric p100 .all 10.0000000 10
tvsports numeric hist .all NA ▇▃▂▂▂▂▂▃
exercise numeric missing .all 36.0000000 36
exercise numeric complete .all 4882.0000000 4882
exercise numeric n .all 4918.0000000 4918
exercise numeric mean .all 6.1175748 6.12
exercise numeric sd .all 2.3290348 2.33
exercise numeric p0 .all 1.0000000 1
exercise numeric p25 .all 5.0000000 5
exercise numeric p50 .all 6.0000000 6
exercise numeric p75 .all 8.0000000 8
exercise numeric p100 .all 10.0000000 10
exercise numeric hist .all NA ▅▃▃▇▇▇▇▇
dining numeric missing .all 36.0000000 36
dining numeric complete .all 4882.0000000 4882
dining numeric n .all 4918.0000000 4918
dining numeric mean .all 7.6874232 7.69
dining numeric sd .all 1.7874685 1.79
dining numeric p0 .all 1.0000000 1
dining numeric p25 .all 7.0000000 7
dining numeric p50 .all 8.0000000 8
dining numeric p75 .all 9.0000000 9
dining numeric p100 .all 10.0000000 10
dining numeric hist .all NA ▁▁▁▂▂▅▅▇
museums numeric missing .all 36.0000000 36
museums numeric complete .all 4882.0000000 4882
museums numeric n .all 4918.0000000 4918
museums numeric mean .all 6.8758705 6.88
museums numeric sd .all 2.0790946 2.08
museums numeric p0 .all 0.0000000 0
museums numeric p25 .all 6.0000000 6
museums numeric p50 .all 7.0000000 7
museums numeric p75 .all 8.0000000 8
museums numeric p100 .all 10.0000000 10
museums numeric hist .all NA ▁▁▂▅▃▇▇▇
art numeric missing .all 36.0000000 36
art numeric complete .all 4882.0000000 4882
art numeric n .all 4918.0000000 4918
art numeric mean .all 6.5948382 6.59
art numeric sd .all 2.2880371 2.29
art numeric p0 .all 0.0000000 0
art numeric p25 .all 5.0000000 5
art numeric p50 .all 7.0000000 7
art numeric p75 .all 8.0000000 8
art numeric p100 .all 10.0000000 10
art numeric hist .all NA ▁▁▃▆▃▆▇▇
hiking numeric missing .all 36.0000000 36
hiking numeric complete .all 4882.0000000 4882
hiking numeric n .all 4918.0000000 4918
hiking numeric mean .all 5.7677181 5.77
hiking numeric sd .all 2.5561746 2.56
hiking numeric p0 .all 0.0000000 0
hiking numeric p25 .all 4.0000000 4
hiking numeric p50 .all 6.0000000 6
hiking numeric p75 .all 8.0000000 8
hiking numeric p100 .all 10.0000000 10
hiking numeric hist .all NA ▂▃▅▇▅▅▆▆
gaming numeric missing .all 36.0000000 36
gaming numeric complete .all 4882.0000000 4882
gaming numeric n .all 4918.0000000 4918
gaming numeric mean .all 4.0202786 4.02
gaming numeric sd .all 2.6734060 2.67
gaming numeric p0 .all 0.0000000 0
gaming numeric p25 .all 2.0000000 2
gaming numeric p50 .all 4.0000000 4
gaming numeric p75 .all 6.0000000 6
gaming numeric p100 .all 14.0000000 14
gaming numeric hist .all NA ▆▇▆▆▂▁▁▁
clubbing numeric missing .all 36.0000000 36
clubbing numeric complete .all 4882.0000000 4882
clubbing numeric n .all 4918.0000000 4918
clubbing numeric mean .all 5.7255223 5.73
clubbing numeric sd .all 2.4457314 2.45
clubbing numeric p0 .all 0.0000000 0
clubbing numeric p25 .all 4.0000000 4
clubbing numeric p50 .all 6.0000000 6
clubbing numeric p75 .all 8.0000000 8
clubbing numeric p100 .all 10.0000000 10
clubbing numeric hist .all NA ▃▂▃▇▅▆▆▅
reading numeric missing .all 36.0000000 36
reading numeric complete .all 4882.0000000 4882
reading numeric n .all 4918.0000000 4918
reading numeric mean .all 7.6448177 7.64
reading numeric sd .all 2.0228217 2.02
reading numeric p0 .all 1.0000000 1
reading numeric p25 .all 7.0000000 7
reading numeric p50 .all 8.0000000 8
reading numeric p75 .all 9.0000000 9
reading numeric p100 .all 13.0000000 13
reading numeric hist .all NA ▁▁▁▅▃▇▁▁
tv numeric missing .all 36.0000000 36
tv numeric complete .all 4882.0000000 4882
tv numeric n .all 4918.0000000 4918
tv numeric mean .all 5.2904547 5.29
tv numeric sd .all 2.4531728 2.45
tv numeric p0 .all 1.0000000 1
tv numeric p25 .all 3.0000000 3
tv numeric p50 .all 6.0000000 6
tv numeric p75 .all 7.0000000 7
tv numeric p100 .all 10.0000000 10
tv numeric hist .all NA ▇▃▃▆▇▆▅▃
theater numeric missing .all 36.0000000 36
theater numeric complete .all 4882.0000000 4882
theater numeric n .all 4918.0000000 4918
theater numeric mean .all 6.7201966 6.72
theater numeric sd .all 2.2523538 2.25
theater numeric p0 .all 0.0000000 0
theater numeric p25 .all 5.0000000 5
theater numeric p50 .all 7.0000000 7
theater numeric p75 .all 8.0000000 8
theater numeric p100 .all 10.0000000 10
theater numeric hist .all NA ▁▁▂▆▃▆▅▇
movies numeric missing .all 36.0000000 36
movies numeric complete .all 4882.0000000 4882
movies numeric n .all 4918.0000000 4918
movies numeric mean .all 7.9799263 7.98
movies numeric sd .all 1.6715580 1.67
movies numeric p0 .all 0.0000000 0
movies numeric p25 .all 7.0000000 7
movies numeric p50 .all 8.0000000 8
movies numeric p75 .all 9.0000000 9
movies numeric p100 .all 10.0000000 10
movies numeric hist .all NA ▁▁▁▁▁▃▅▇
concerts numeric missing .all 36.0000000 36
concerts numeric complete .all 4882.0000000 4882
concerts numeric n .all 4918.0000000 4918
concerts numeric mean .all 6.8244572 6.82
concerts numeric sd .all 2.0960622 2.1
concerts numeric p0 .all 0.0000000 0
concerts numeric p25 .all 6.0000000 6
concerts numeric p50 .all 7.0000000 7
concerts numeric p75 .all 8.0000000 8
concerts numeric p100 .all 10.0000000 10
concerts numeric hist .all NA ▁▁▂▅▆▆▇▇
music numeric missing .all 36.0000000 36
music numeric complete .all 4882.0000000 4882
music numeric n .all 4918.0000000 4918
music numeric mean .all 7.7808275 7.78
music numeric sd .all 1.8393525 1.84
music numeric p0 .all 1.0000000 1
music numeric p25 .all 7.0000000 7
music numeric p50 .all 8.0000000 8
music numeric p75 .all 9.0000000 9
music numeric p100 .all 10.0000000 10
music numeric hist .all NA ▁▁▁▂▂▃▅▇
shopping numeric missing .all 36.0000000 36
shopping numeric complete .all 4882.0000000 4882
shopping numeric n .all 4918.0000000 4918
shopping numeric mean .all 5.4840229 5.48
shopping numeric sd .all 2.5703482 2.57
shopping numeric p0 .all 1.0000000 1
shopping numeric p25 .all 3.0000000 3
shopping numeric p50 .all 6.0000000 6
shopping numeric p75 .all 7.0000000 7
shopping numeric p100 .all 10.0000000 10
shopping numeric hist .all NA ▇▃▅▆▆▆▅▆
yoga numeric missing .all 36.0000000 36
yoga numeric complete .all 4882.0000000 4882
yoga numeric n .all 4918.0000000 4918
yoga numeric mean .all 4.2115936 4.21
yoga numeric sd .all 2.7052281 2.71
yoga numeric p0 .all 0.0000000 0
yoga numeric p25 .all 2.0000000 2
yoga numeric p50 .all 4.0000000 4
yoga numeric p75 .all 6.0000000 6
yoga numeric p100 .all 10.0000000 10
yoga numeric hist .all NA ▇▆▆▇▅▃▂▃
attr numeric missing .all 118.0000000 118
attr numeric complete .all 4800.0000000 4800
attr numeric n .all 4918.0000000 4918
attr numeric mean .all 6.0637500 6.06
attr numeric sd .all 1.9491874 1.95
attr numeric p0 .all 0.0000000 0
attr numeric p25 .all 5.0000000 5
attr numeric p50 .all 6.0000000 6
attr numeric p75 .all 7.0000000 7
attr numeric p100 .all 10.0000000 10
attr numeric hist .all NA ▁▁▂▇▆▆▅▃
sinc numeric missing .all 161.0000000 161
sinc numeric complete .all 4757.0000000 4757
sinc numeric n .all 4918.0000000 4918
sinc numeric mean .all 7.0538154 7.05
sinc numeric sd .all 1.8065752 1.81
sinc numeric p0 .all 0.0000000 0
sinc numeric p25 .all 6.0000000 6
sinc numeric p50 .all 7.0000000 7
sinc numeric p75 .all 8.0000000 8
sinc numeric p100 .all 10.0000000 10
sinc numeric hist .all NA ▁▁▁▅▆▇▇▆
intel numeric missing .all 166.0000000 166
intel numeric complete .all 4752.0000000 4752
intel numeric n .all 4918.0000000 4918
intel numeric mean .all 7.2659933 7.27
intel numeric sd .all 1.5855446 1.59
intel numeric p0 .all 0.0000000 0
intel numeric p25 .all 6.0000000 6
intel numeric p50 .all 7.0000000 7
intel numeric p75 .all 8.0000000 8
intel numeric p100 .all 10.0000000 10
intel numeric hist .all NA ▁▁▁▃▅▇▇▆
fun numeric missing .all 197.0000000 197
fun numeric complete .all 4721.0000000 4721
fun numeric n .all 4918.0000000 4918
fun numeric mean .all 6.2887100 6.29
fun numeric sd .all 1.9761549 1.98
fun numeric p0 .all 0.0000000 0
fun numeric p25 .all 5.0000000 5
fun numeric p50 .all 6.0000000 6
fun numeric p75 .all 8.0000000 8
fun numeric p100 .all 10.0000000 10
fun numeric hist .all NA ▁▁▂▇▇▇▆▃
amb numeric missing .all 421.0000000 421
amb numeric complete .all 4497.0000000 4497
amb numeric n .all 4918.0000000 4918
amb numeric mean .all 6.6965755 6.7
amb numeric sd .all 1.8329494 1.83
amb numeric p0 .all 0.0000000 0
amb numeric p25 .all 6.0000000 6
amb numeric p50 .all 7.0000000 7
amb numeric p75 .all 8.0000000 8
amb numeric p100 .all 10.0000000 10
amb numeric hist .all NA ▁▁▁▇▇▇▇▆
shar numeric missing .all 643.0000000 643
shar numeric complete .all 4275.0000000 4275
shar numeric n .all 4918.0000000 4918
shar numeric mean .all 5.3198830 5.32
shar numeric sd .all 2.1648979 2.16
shar numeric p0 .all 0.0000000 0
shar numeric p25 .all 4.0000000 4
shar numeric p50 .all 5.0000000 5
shar numeric p75 .all 7.0000000 7
shar numeric p100 .all 10.0000000 10
shar numeric hist .all NA ▁▂▂▇▅▃▂▂
like numeric missing .all 122.0000000 122
like numeric complete .all 4796.0000000 4796
like numeric n .all 4918.0000000 4918
like numeric mean .all 6.0513970 6.05
like numeric sd .all 1.8513350 1.85
like numeric p0 .all 0.0000000 0
like numeric p25 .all 5.0000000 5
like numeric p50 .all 6.0000000 6
like numeric p75 .all 7.0000000 7
like numeric p100 .all 10.0000000 10
like numeric hist .all NA ▁▁▂▇▇▇▅▂
prob numeric missing .all 156.0000000 156
prob numeric complete .all 4762.0000000 4762
prob numeric n .all 4918.0000000 4918
prob numeric mean .all 5.0170097 5.02
prob numeric sd .all 2.1651088 2.17
prob numeric p0 .all 0.0000000 0
prob numeric p25 .all 4.0000000 4
prob numeric p50 .all 5.0000000 5
prob numeric p75 .all 7.0000000 7
prob numeric p100 .all 10.0000000 10
prob numeric hist .all NA ▂▂▂▇▃▃▂▁
match_es numeric missing .all 460.0000000 460
match_es numeric complete .all 4458.0000000 4458
match_es numeric n .all 4918.0000000 4918
match_es numeric mean .all 3.1689771 3.17
match_es numeric sd .all 2.3628158 2.36
match_es numeric p0 .all 0.0000000 0
match_es numeric p25 .all 2.0000000 2
match_es numeric p50 .all 3.0000000 3
match_es numeric p75 .all 4.0000000 4
match_es numeric p100 .all 10.0000000 10
match_es numeric hist .all NA ▇▇▆▇▁▁▁▂
attr3_s logical missing .all 4918.0000000 4918
attr3_s logical complete .all 0.0000000 0
attr3_s logical n .all 4918.0000000 4918
attr3_s logical mean .all NaN NaN
attr3_s logical count NA 4918.0000000 4918
sinc3_s logical missing .all 4918.0000000 4918
sinc3_s logical complete .all 0.0000000 0
sinc3_s logical n .all 4918.0000000 4918
sinc3_s logical mean .all NaN NaN
sinc3_s logical count NA 4918.0000000 4918
intel3_s logical missing .all 4918.0000000 4918
intel3_s logical complete .all 0.0000000 0
intel3_s logical n .all 4918.0000000 4918
intel3_s logical mean .all NaN NaN
intel3_s logical count NA 4918.0000000 4918
fun3_s logical missing .all 4918.0000000 4918
fun3_s logical complete .all 0.0000000 0
fun3_s logical n .all 4918.0000000 4918
fun3_s logical mean .all NaN NaN
fun3_s logical count NA 4918.0000000 4918
amb3_s logical missing .all 4918.0000000 4918
amb3_s logical complete .all 0.0000000 0
amb3_s logical n .all 4918.0000000 4918
amb3_s logical mean .all NaN NaN
amb3_s logical count NA 4918.0000000 4918
dec character missing .all 0.0000000 0
dec character complete .all 4918.0000000 4918
dec character n .all 4918.0000000 4918
dec character min .all 2.0000000 2
dec character max .all 3.0000000 3
dec character empty .all 0.0000000 0
dec character n_unique .all 2.0000000 2
match numeric missing .all 0.0000000 0
match numeric complete .all 4918.0000000 4918
match numeric n .all 4918.0000000 4918
match numeric mean .all 0.4158194 0.42
match numeric sd .all 0.4929128 0.49
match numeric p0 .all 0.0000000 0
match numeric p25 .all 0.0000000 0
match numeric p50 .all 0.0000000 0
match numeric p75 .all 1.0000000 1
match numeric p100 .all 1.0000000 1
match numeric hist .all NA ▇▁▁▁▁▁▁▆

As seguintes variáveis irão ser utilizadas na regressão:

mod <- glm(match ~ prob+attr+like+int_corr+intel,
      data = data,  
      family = "binomial")

tidy(mod, conf.int = TRUE, exponentiate = TRUE)
term estimate std.error statistic p.value conf.low conf.high
(Intercept) 0.0014839 0.2548753 -25.5540040 0.0000000 0.0008942 0.0024290
prob 1.1992670 0.0202251 8.9844159 0.0000000 1.1528091 1.2479521
attr 1.5313203 0.0279984 15.2197941 0.0000000 1.4501031 1.6183677
like 1.9185639 0.0363641 17.9181153 0.0000000 1.7879198 2.0619117
int_corr 1.0380300 0.1211887 0.3079885 0.7580911 0.8186087 1.3165521
intel 0.8159728 0.0302849 -6.7153597 0.0000000 0.7687265 0.8656521
glance(mod)
null.deviance df.null logLik AIC BIC deviance df.residual
6331.339 4637 -2208.293 4428.587 4467.239 4416.587 4632
pR2(mod)
##           llh       llhNull            G2      McFadden          r2ML 
## -2208.2932637 -3338.8632511  2261.1399748     0.3386093     0.3858553 
##          r2CU 
##     0.5056940

Abaixo teremos um gráfico que mostra o coeficiente esperado para cada umas das variáveis escolhidas para fazer parte do modelo.

tidy(mod, conf.int = TRUE, conf.level = 0.95, exponentiate = TRUE) %>%
  filter(term != "(Intercept)") %>%
  ggplot(aes(term, estimate, ymin = conf.low, ymax = conf.high)) +
  geom_bar(stat = "identity") + 
  geom_hline(yintercept = 1, colour = "darkred") +
  labs(x = "Variáveis Analisadas",
       title = "Regressão Logística (Intervalo)",
       y = expression("Coeficientes"))

Pergunta a ser respondida: Que fatores nos dados têm efeito relevante na chance do casal ter um match? Descreva se os efeitos são positivos ou negativos e sua magnitude.

Ao ser utilizado a regressão múltipla logística usando as variáveis

chegamos ao seguinte modelo:

Tal modelo explica 33.86% da variância da variável resposta segundo McFadden.

Ao analisarmos o modelo podemos concluir que as seguintes variáveis possuem efeito relevante na chance de um casal dar “match” (IC 95%):

As demais variáveis (menos relevantes) são descritas a seguir (IC 95%):